Storage Subsystem Administration Guide - Hitachi Data … · Storage Subsystem Administration Guide...

Storage Subsystem Administration GuideHitachi Unified NAS Module

Hitachi NAS Platform

Release 12.7

MK-92HNAS012-09

July 2016

© 2011-2016 Hitachi, Ltd. All rights reserved.

No part of this publication may be reproduced or transmitted in any form orby any means, electronic or mechanical, including photocopying andrecording, or stored in a database or retrieval system for any purposewithout the express written permission of Hitachi, Ltd.

Hitachi, Ltd., reserves the right to make changes to this document at anytime without notice and assumes no responsibility for its use. This documentcontains the most current information available at the time of publication.When new or revised information becomes available, this entire documentwill be updated and distributed to all registered users.

Some of the features described in this document might not be currentlyavailable. Refer to the most recent product announcement for informationabout feature and product availability, or contact Hitachi Data SystemsCorporation at https://portal.hds.com.

Notice: Hitachi, Ltd., products and services can be ordered only under theterms and conditions of the applicable Hitachi Data Systems Corporationagreements. The use of Hitachi, Ltd., products is governed by the terms ofyour agreements with Hitachi Data Systems Corporation.

2Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

https://portal.hds.com

Hitachi Data Systems products and services can be ordered only under the terms andconditions of Hitachi Data Systems’ applicable agreements. The use of Hitachi DataSystems products is governed by the terms of your agreements with Hitachi DataSystems.

By using this software, you agree that you are responsible for:

a) Acquiring the relevant consents as may be required under local privacy laws orotherwise from employees and other individuals to access relevant data; and

b) Verifying that data continues to be held, retrieved, deleted, or otherwise processedin accordance with relevant laws.

Hitachi is a registered trademark of Hitachi, Ltd., in the United States and othercountries. Hitachi Data Systems is a registered trademark and service mark ofHitachi, Ltd., in the United States and other countries.

Archivas, Dynamic Provisioning, Essential NAS Platform, HiCommand, Hi-Track,ShadowImage, Tagmaserve, Tagmasoft, Tagmasolve, Tagmastore, TrueCopy,Universal Star Network, and Universal Storage Platform are registered trademarks ofHitachi Data Systems Corporation.

AIX, AS/400, DB2, Domino, DS8000, Enterprise Storage Server, ESCON, FICON,FlashCopy, IBM, Lotus, OS/390, RS6000, S/390, System z9, System z10, Tivoli, VM/ESA, z/OS, z9, zSeries, z/VM, z/VSE are registered trademarks and DS6000, MVS,and z10 are trademarks of International Business Machines Corporation.

All other trademarks, service marks, and company names in this document orwebsite are properties of their respective owners.

Microsoft product screen shots are reprinted with permission from MicrosoftCorporation.

This product includes software developed by the OpenSSL Project for use in theOpenSSL Toolkit (http://www.openssl.org/). Some parts of ADC use open source codefrom Network Appliance, Inc. and Traakan, Inc.

Part of the software embedded in this product is gSOAP software. Portions created bygSOAP are copyright 2001-2009 Robert A. Van Engelen, Genivia Inc. All rightsreserved. The software in this product was in part provided by Genivia Inc. and anyexpress or implied warranties, including, but not limited to, the implied warranties ofmerchantability and fitness for a particular purpose are disclaimed. In no event shallthe author be liable for any direct, indirect, incidental, special, exemplary, orconsequential damages (including, but not limited to, procurement of substitutegoods or services; loss of use, data, or profits; or business interruption) howevercaused and on any theory of liability, whether in contract, strict liability, or tort(including negligence or otherwise) arising in any way out of the use of this software,even if advised of the possibility of such damage.

The product described in this guide may be protected by one or more U.S. patents,foreign patents, or pending applications.


Notice of Export Controls

Export of technical data contained in this document may require an exportlicense from the United States government and/or the government of Japan.Contact the Hitachi Data Systems Legal Department for any exportcompliance questions.



Contents

Preface ................................................................................................ 8Related Documentation............................................................................................ 8Accessing product documentation........................................................................... 11Getting help...........................................................................................................11Comments.............................................................................................................11

1 Understanding storage and tiering....................................................12Understanding tiered storage.................................................................................. 13Storage management components.......................................................................... 13

System drives...................................................................................................14Creating system drives................................................................................ 14

Storage pools................................................................................................... 14Tiered storage pools....................................................................................15Storage pool chunks....................................................................................16

Dynamically provisioned volumes.......................................................................17Dynamically provisioned pools........................................................................... 17File system types..............................................................................................18Fibre Channel connections.................................................................................18

About FC paths........................................................................................... 19Load balancing and failure recovery..............................................................20Fibre channel statistics................................................................................ 21

RAID controllers............................................................................................... 22

2 Managing the storage subsystem..................................................... 24Supported Hitachi Data Systems storage subsystems................................................25Using Hitachi Dynamic Provisioning......................................................................... 25

HDP high-level process......................................................................................25Understanding HDP thin provisioning................................................................. 26Thin provisioning using the NAS server filesystem-thin command......................... 28Deciding how far to over-provision storage......................................................... 29Understanding how HDP works with HNAS......................................................... 30


Configuring storage for HDP and HNAS.............................................................. 30Configuring storage to use HDP....................................................................31

Configuration considerations when using HDP with HNAS ................................... 32Using HDP pools with an HNAS server.......................................................... 33Configuration guidelines for HNAS with HDP..................................................33Configuring storage pool pre-allocation......................................................... 34Performing large file system expansions........................................................36Upgrading from older HNAS systems............................................................ 36

Using HDP storage............................................................................................37Considerations when using HDP pools...........................................................37Creating an HDP pool with untiered storage.................................................. 37Creating HDP pools with tiered storage.........................................................37Creating storage pools with DP pools from HDP storage.................................38Moving free space between storage pools..................................................... 38

System drive groups and dynamic write balancing.................................................... 41Read balancing utility considerations....................................................................... 43Snapshots and the file system data redistribution utility............................................ 44

3 Using a storage pool........................................................................46Creating storage pools............................................................................................47

Creating a storage pool using the GUI................................................................47Creating a storage pool using the CLI.................................................................50

Adding the metadata tier........................................................................................ 51Expanding storage pools.........................................................................................52

Why use HDP to expand DP-Vols........................................................................52Expanding space in a thinly provisioned HDP storage pool..............................53Expanding storage space using DP-Vols........................................................ 54

Expanding a non-HDP storage pool or tier.......................................................... 54Configuring automatic file system expansion for an entire storage pool...................... 56Renaming a storage pool........................................................................................ 58

Storage pool naming.........................................................................................58Reducing the size of a storage pool......................................................................... 60Denying access to a storage pool............................................................................ 60Allowing access to a storage pool............................................................................ 62Deleting a storage pool...........................................................................................62


Preface

In PDF format, this guide provides information about managing the supportedstorage subsystems attached to the server/cluster. The guide includesinformation about tiered storage, storage pools, system drives (SDs), usingHitachi Dynamic Provisioning (HDP) with the server, and other storage relatedconfiguration and management features and functions.

Related DocumentationRelease Notes provide the most up-to-date information about the system,including new feature summaries, upgrade instructions, and fixed and knowndefects.

Command Line ReferencesThe Command Line Reference provides information on the commands used tomanage your system, and includes relevant information on the operation ofyour hardware and software. Depending on the model of your server orcluster node, refer to the Command Line Reference that is appropriate foryour system.• NAS Module Server Command Line Reference• Command Line Reference for models 4060, 4080, and 4100• Command Line Reference for models 3080 and 3090

Administration Guides

• System Access Guide (MK-92HNAS014)—Explains how to log in to thesystem, provides information about accessing the NAS server/cluster CLIand the SMU CLI, and provides information about the documentation, help,and search capabilities available in the system.

• Server and Cluster Administration Guide (MK-92HNAS010)—Providesinformation about administering servers, clusters, and server farms.Includes information about licensing, name spaces, upgrading firmware,monitoring servers and clusters, the backing up and restoringconfigurations.

• Storage System User Administration Guide (MK-92HNAS013)—Explainsuser management, including the different types of system administrator,their roles, and how to create and manage these users.

• Network Administration Guide (MK-92HNAS008)—Provides informationabout the server's network usage, and explains how to configure networkinterfaces, IP addressing, name and directory services.

• File Services Administration Guide (MK-92HNAS006)—Explains about filesystem formats, and provides information about creating and managing

8 PrefaceHitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

file systems, and enabling and configuring file services (file serviceprotocols).

• Data Migrator Administration Guide (MK-92HNAS005) —Providesinformation about the Data Migrator feature, including how to set upmigration policies and schedules.

• Storage Subsystem Administration Guide (MK-92HNAS012)—Providesinformation about managing the supported storage subsystems (RAIDarrays) attached to the server/cluster. Includes information about tieredstorage, storage pools, system drives (SDs), SD groups, and other storagedevice related configuration and management features and functions.

• Snapshot Administration Guide (MK-92HNAS011)—Provides informationabout configuring the server to take and manage snapshots.

• Replication and Disaster Recovery Administration Guide (MK-92HNAS009)—Provides information about replicating data using file-based replicationand object-based replication, provides information on setting up replicationpolicies and schedules, and using replication features for disaster recoverypurposes.

• Antivirus Administration Guide (MK-92HNAS004)—Describes the supportedantivirus engines, provides information about how to enable them, andhow to configure the system to use them.

• Backup Administration Guide (MK-92HNAS007)—Provides informationabout configuring the server to work with NDMP, and making andmanaging NDMP backups.

Note: For a complete list of Hitachi NAS open source software copyrights andlicenses, see the System Access Guide.

Hardware References• Hitachi NAS Platform 3080 and 3090 G2 Hardware Reference

(MK-92HNAS017) —Provides an overview of the second-generation serverhardware, describes how to resolve any problems, and replace potentiallyfaulty parts.

• Hitachi NAS Platform and Hitachi Unified Storage Series 4000 HardwareReference (MK-92HNAS030)—Provides an overview of the Hitachi NASPlatform Series 4000 server hardware, describes how to resolve anyproblems, and how to replace potentially faulty components

• Hitachi NAS Platform System Manager Unit (SMU) Hardware Reference(MK-92HNAS065)—This document describes the usage and replacementinstructions for the SMU 300/400.

Best Practices• Hitachi USP-V/VSP Best Practice Guide for HNAS Solutions

(MK-92HNAS025)—The practices outlined in this document describe how toconfigure the system to achieve the best results.

• Hitachi Unified Storage VM Best Practices Guide for HNAS Solutions(MK-92HNAS026) —The system is capable of heavily driving a storage

Preface 9Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

array and disks. The practices outlined in this document describe how toconfigure the system to achieve the best results

• Hitachi NAS Platform Best Practices Guide for NFS with VMware vSphere(MK-92HNAS028)—This document covers best practices specific to usingVMware vSphere with the Hitachi NAS platform.

• Hitachi NAS Platform Deduplication Best Practice (MK-92HNAS031)—Thisdocument provides best practices and guidelines for using deduplication.

• Hitachi NAS Platform Best Practices for Tiered File Systems(MK-92HNAS038)—This document describes the Hitachi NAS Platformfeature that automatically and intelligently separates data and metadataonto different Tiers of storage called Tiered File Systems (TFS).

• Hitachi NAS Platform Data Migrator to Cloud Best Practices Guide(MK-92HNAS045)—Data Migrator to Cloud allows files hosted on the HNASserver to be transparently migrated to cloud storage, providing thebenefits associated with both local and cloud storage.

• Brocade VDX 6730 Switch Configuration for use in an HNAS ClusterConfiguration Guide (MK-92HNAS046)—This document describes how toconfigure a Brocade VDX 6730 switch for use as an ISL (inter-switch link)or an ICC (inter-cluster communication) switch.

• Best Practices for Hitachi NAS Universal Migrator (MK-92HNAS047)—TheHitachi NAS Universal Migrator (UM) feature provides customers with aconvenient and minimally disruptive method to migrate from their existingNAS system to the Hitachi NAS Platform. The practices andrecommendations outlined in this document describe how to best use thisfeature.

• Hitachi Data Systems SU 12.x Network File System (NFS) Version 4Feature Description (MK-92HNAS056)—This document describes thefeatures of Network File System (NFS) Version 4.

• Hitachi NAS 12.1 HDP Best Practices (MK-92HNAS057)—This documentlists frequently asked questions regarding the use of Hitachi DynamicProvisioning.

• Hitachi Multi-tenancy Implementation and Best Practice Guide(MK-92HNAS059)—This document details the best practices for configuringand using Multi-Tenancy and related features, and EVS security.

• Hitachi NAS Platform v 12.1 HDP Best Practices (MK-92HNAS063)—Thisdocument details the best practices for configuring and using storagepools, related features, and Hitachi Dynamic Provisioning (HDP).

• Hitachi NAS Platform System Manager Unit (SMU) Hardware Reference(MK-92HNAS065)—This document describes the usage and replacementinstructions for the SMU 300/400. —

• Brocade VDX 6740 Switch Configuration for use in an HNAS ClusterConfiguration Guide (MK-92HNAS066)—This document describes how toconfigure a Brocade VDX 6740 switch for use as an ICC (intra-clustercommunication) switch.

• File System Snapshots Operational Best Practice (MK-92HNAS068)—Thisdocument provides operational guidance on file system snapshots.

10 PrefaceHitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

• Virtual Infrastructure Integrator for Hitachi Storage Platforms OperationalBest Practice (MK-92HNAS069)—This document provides operationalguidance on Hitachi Virtual Infrastructure Integrator for the HNASplatform.

• Hitachi NAS Platform Replication Best Practices Guide (MK-92HNAS070)—This document details the best practices for configuring and using HNASReplication and related features.

Accessing product documentationProduct user documentation is available on Hitachi Data Systems SupportConnect: https://support.hds.com/en_us/documents.html. Check this site forthe most current documentation, including important updates that may havebeen made after the release of the product.

Getting helpHitachi Data Systems Support Connect is the destination for technical supportof products and solutions sold by Hitachi Data Systems. To contact technicalsupport, log on to Hitachi Data Systems Support Connect for contactinformation: https://support.hds.com/en_us/contact-us.html.

Hitachi Data Systems Community is a global online community for HDScustomers, partners, independent software vendors, employees, andprospects. It is the destination to get answers, discover insights, and makeconnections. Join the conversation today! Go to community.hds.com,register, and complete your profile.

CommentsPlease send us your comments on this document to [email protected] the document title and number, including the revision level (forexample, -07), and refer to specific sections and paragraphs wheneverpossible. All comments become the property of Hitachi Data SystemsCorporation.

Thank you!

Preface 11Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

https://support.hds.com/en_us/documents.html

https://support.hds.com/en_us/contact-us.html

https://support.hds.com/en_us/contact-us.html

https://community.hds.com/welcome

https://community.hds.com/welcome

mailto:[email protected]

1Understanding storage and tiering

□ Understanding tiered storage

□ Storage management components

12 Understanding storage and tieringHitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

Understanding tiered storageTiered storage allows you to connect multiple diverse storage subsystemsbehind a single server (or cluster). Using tiered storage, you can matchapplication storage requirements (in terms of performance and scaling) toyour storage subsystems. This section describes the concept of tieredstorage, and explains how to configure the storage server to work with yourstorage subsystems to create a tiered storage architecture.

Based on a storage subsystem's performance characteristics, it is classifiedas belonging to a certain tier, and each tier is used differently in theenterprise storage architecture. The currently supported storage subsystemsfit into the tiered storage model as follows:

Tier Performance Drive Type RPM (ifdisk)

0 Extremely high Flash or solid-state memory; not disk N/A

1 Very high SAS 15,000

2 High SAS 10,000

3 Nearline Nearline (NL) SAS 7,200

4 Archival Nearline (NL) SAS 7,200

5 Long-term storage (Tape) N/A (Tape) N/A

The NAS server supports tiers of storage, where each tier is made up ofdevices with different performance characteristics or technologies. The NASserver also supports storage virtualization through Hitachi Universal StoragePlatform VSP, USP-V, USP-VM, and HUS-VM technology.

Tiers of storage and storage virtualization are fully supported by DataMigrator, an optional feature which allows you to optimize the usage of tieredstorage and remote NFSv3 servers (note, however that Data Migrator doesnot support migration to or from tape storage devices or tape librarysystems). For detailed information about Data Migrator, refer to the DataMigrator Administration Guide.

Storage management componentsThe storage server architecture includes system drives, storage pools, filesystems and virtual servers (EVSs), supplemented by a flexible quotamanagement system for managing utilization, and the Data Migrator, whichoptimizes available storage. This section describes each of these storagecomponents and functions in detail.

Understanding storage and tiering 13Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

System drivesSystem drives (SDs) are the basic logical storage element used by the server.Storage subsystems use RAID controllers to aggregate multiple physical disksinto SDs (also known as LUs). An SD is a logical unit made up of made up ofa group of physical disks or flash/SSD drives. The size of the SD depends onfactors such as the RAID level, the number of drives, and their capacity.

Some Hitachi Data Systems RAID arrays, such as HUS VM, have a limit of 3TiB for standard LUs or 4 TiB for virtualized LUs. When using legacy storagearrays, it is a common practice for system administrators to build large RAIDgroups (often called parity groups or volume groups) and then divide theminto SDs (LUs) of 2 TiB or less. With today's large physical disks, RAID groupsmust be considerably larger than 2 TiB in order to make efficient use ofspace.

Creating system drives

When creating SDs, use the Hitachi storage management applicationappropriate for your storage subsystem. You cannot create SDs using NASManager or the NAS server command line.

When creating SDs, you may need to specify array-specific settings in thestorage management application. Also, depending on the firmware version ofthe array, there may be device-specific configuration settings. For example,on HUS 110, HUS 130, and HUS 150 arrays, if the HUS1x0 firmware code isbase 0935A or greater, you should enable the HNAS Option Mode on theOptions tab of the Edit Host Groups page.

For more information about what settings are required for each type of array,and for the firmware installed on the array, contact Hitachi Data Systemscustomer support.

Storage poolsA NAS server storage pool (known as a "span" in the command line interface)is the logical container for a collection of four or more system drives (SDs).There are two types of NAS server storage pools:• An untiered storage pool is made up system drives (SDs) created on one

or more storage subsystems within the same tier of storage (storagesubsystems with comparable performance characteristics). To create anuntiered storage pool, there must be at least four available and unusedsystem drives on the storage subsystem from which the SDs in the storagepool will be taken.

• A tiered storage pool is made up system drives (SDs) created on storagesubsystems with different performance characteristics. Typically, a tieredstorage pool is made up of SDs from high-performance storage such asSSD/flash memory, and SDs from lower-performance storage such as SAS,


or NL SAS (near line SAS). You can, however, create a tiered storage poolfrom SDs on storage subsystems using any storage technology, and youcan create both tiers on the same storage subsystem.

NAS server storage pools:• Can be expanded as additional SDs are created in the storage subsystem,

and a storage pool can grow to a maximum of 1 PiB or 256 SDs.Expanding a NAS server storage pool does not interrupt network clientaccess to storage resources. SDs may be based on parity groups, or onHDP DP-Vols.

• Support two types of thin provisioning:○ NAS server storage pools can be thinly provisioned when created using

SDs based on HDP DP-Vols.○ File system thin provisioning, through the use of the NAS server

filesystem-thin command and file system confinement and auto-expansion. This type of thin provisioning allows you to create a smallfile system, which can then automatically expand when necessary, andthat ability saves the overhead associated with sustaining unnecessarystorage.When file system thin provisioning is enabled, the server reports toprotocol clients (though not at the CLI or in the GUI) that the filesystem is larger than it really is: either the capacity to which you haveconfined it or the maximum capacity to which it can ever grow,whichever is smaller.Refer to the Command Line Reference for more information on thefilesystem-thin command.

• Contain a single stripeset when created. Each time the storage pool isexpanded, another stripeset is added, up to a maximum of 64 stripesets(meaning that, after creation, a storage pool can be expanded a maximumof 63 times).

• Contain the file systems, centralizing and simplifying the management offile system settings that are common to all file systems in the storagepool. For example, the settings applied to a storage pool can either allowor constrain the expansion of all file systems in the storage pool.

Note: The recommended best practice for the number of filesystems per storage pool is that there should be no more than 20file systems per storage pool. Recently deleted file systems that arestill in the recycle bin do not count towards this number.

Tiered storage pools

Currently, a tiered storage pool must have two tiers:• Tier 0 is used for metadata, and the best-performing storage should be

designated as Tier 0.• Tier 1 is used for user data.


When creating a tiered storage pool, at least four unused SDs must beavailable for each tier. When you create a tiered storage pool, you first createthe user data tier (Tier 1), then you create the metadata tier (Tier 0).

During normal operation, one tier of a tiered storage pool might become filledbefore the other tier. In such a case, you can expand one tier of the storagepool without expanding the other tier. When expanding a tier, you must:• Make certain that the SDs being added to the tier have the same

performance characteristics as the SDs already in the tier (for example, donot add NL SAS (near line SAS) based SDs to a tier already made up ofSSD/flash drives).

• Add at least four SDs to the tier.

See the span-create man page for more information about creating storagepools.

Storage pool chunksThis section explains what storage pool chunks are, and how they affectstorage pool and file system size and performance.

Storage pools are made up of multiple small allocations of storage called“chunks.” The size of the chunks in a storage pool is defined when thestorage pool is created, and a guideline chunk size is set. The guideline chunksize determines the maximum size that the storage pool (span) can everreach, because a storage pool can contain up to a maximum of 60,000chunks. The guideline chunk size is between 500 MiB and 18 GiB.

Chunk size is an important consideration when creating storage pools, for tworeasons:• Chunks define the increment by which file systems will grow when they

expand. Smaller chunks increase storage efficiency and the predictabilityof expansions, smaller chunks also work better with tiered file systems,and they also enable the effective optimization of I/O patterns byspreading I/O across multiple stripesets.

• An individual file system can contain up to a maximum of 60,000 chunks(the same number as a storage pool). As a file system contains a finitenumber of chunks, the chunk size places a limit on the future growth of filesystems in a storage pool.○ A smaller chunk size will result in a storage pool that expands in a more

granular fashion, but cannot expand to as large an overall size as astorage pool created using the a large default chunk size (for example,an 18GiB chunk size).

○ A larger chunk size will result in storage pools that can expand to alarger overall size than those created using a smaller chunk size, butthe storage pool will expand in larger increments.

The guideline chunk size is specified when creating the storage pool:


• If you create a storage pool using the CLI, the server will calculate aguideline chunk size based on the initial size specified for the storage pooldivided by 3750, but with a minimum of 1 GiB. The server-calculatedguideline chunk size will probably be smaller than NAS Manager would use(NAS Manager will always use a guideline chunk size of 18 GiB).

• If you create a storage pool using NAS Manager, the guideline chunk sizewill be 18 GiB (the maximum allowable size). The default chunk size set byNAS Manager may (and probably will) be larger than the guideline chunksize calculated and suggested by the server if you created the storage poolusing the CLI.

Note: When creating a storage pool using the HNAS server CLI, you canspecify to use a guideline chunk size other than what the server calculates.When creating a storage pool using NAS Manager, you cannot change theguideline chunk size used when creating a storage pool.

Dynamically provisioned volumesA dynamically provisioned volume (DP-Vol) is a virtualized logical unit (LU)that is used with Hitachi Dynamic Provisioning (HDP). You create DP-Vols in adynamically provisioned pool.

The total capacity of a DP-Vol can exceed that of the underlying parity groupsor pool volumes (called thin provisioning). Every DP-Vol can draw space fromany of the underlying parity groups or pool volumes, so the DP-Vol performswell, even during periods of heavy load and high activity (hot spots).

Hitachi Dynamic Provisioning (HDP) thin provisioning enables granular span-expansion without loss of performance, which is impossible without dynamicprovisioning. The NAS server is aware of thin provisioning, and does not usemore space than actually exists, making thin provisioning quite safe with aNAS server. Unlike other server platforms, when HDP thin provisioning isused, the NAS server does not recommend that the DP pool be expandedwhen it becomes 70% full.

Dynamically provisioned poolsA dynamically provisioned pool (DP pool) is an expandable collection of paritygroups or pool volumes containing the dynamically provisioned volume (DP-Vols). DP pools are also sometimes referred to as HDP pools.

On enterprise storage, a DP pool resides on the pool volumes. On modularstorage, a DP pool resides on the parity groups (PGs), rather than on logicalunits (LUs).

Note: Real (non-virtual) LUs are referred to as pool volumes in enterprisestorage. In modular storage, real LUs are referred to a parity groups.


File system typesA file system typically consists of files and directories. Data about the filesand directories (as well as many other attributes) is the metadata. The datawithin the file system (both user data and metadata) is stored in a storagepool.

Like storage pools, file system data (metadata and user data) may be storedin a single tier, or in multiple tiers.• When file system metadata and user data are stored on storage

subsystems of a single storage tier, the file system is called an untiered filesystem. An untiered file system must be created in an untiered storagepool, it cannot be created in a tiered storage pool.

• When file system metadata and user data are stored on storagesubsystems of different storage tiers, the file system is called a tiered filesystem.In a tiered file system, metadata is stored on the highest performance tierof storage, and user data is stored on a lower-performance tier. Storingmetadata on the higher-performance tier provides system performancebenefits over storing both the metadata and user data on the same tier ofstorage.A tiered file system must be created in a tiered storage pool; it cannot becreated in an untiered storage pool.

Fibre Channel connections

Note: The number and operational speed of fibre channel ports on a NASserver are dependent on the server model. Refer to the hardware manual foryour server model for more information on the number, operational speed,and location of fibre channel ports on your NAS server.

Unified NAS Module servers

Depending on the model of your Unified NAS Module server, it may contain 8Gbps and/or 16 Gbps fibre channel ports. Refer to the Hardware ReferenceGuide for your system to find more information about the fibre channel portson your Unified NAS Module system. Contact Hitachi Data Systems customersupport for information about using these ports.

Hitachi NAS Platform servers

Each Hitachi NAS Platform server supports up to four independentlyconfigurable FC ports. Independent configuration allows you to connect to arange of storage subsystems, which allows you to choose the configurationthat will best meet application requirements. The server manages all back-


end storage as a single system, through an integrated network managementinterface.

Hitachi NAS Platform server model Supported FC port operational speeds

3080, 3090, 3100, and 4040 1, 2, or 4 Gbps

4060, 4080, and 4100 2, 4, or 8 Gbps

The server supports connecting to storage arrays either through direct-attached FC connections to the storage array (also called DAS connections) orFibre Channel switches connected to the storage array (also called SANconfigurations):• In direct-attached (DAS) configurations, you can connect up to two (2)

storage arrays directly to a server or a two-node cluster. Clusters of morethan two nodes must use a FC switch configuration.

• In configurations using FC switches (SAN configurations), the server mustbe configured for N_Port operation. Several FC Switch options areavailable, contact Hitachi Data Systems customer support for moreinformation.

You can manage the FC interface on the server/cluster through the commandline interface (CLI), using the following commands:• fc-link to enable or disable the FC link.• fc-link-type to change the FC link type.• fc-link-speed to change the FC interface speed.

For more information about these commands, refer to the Command LineReference.

About FC paths

The NAS server accesses the storage subsystem through a minimum of twoFC paths (at least one from each of the Fibre Channel switches). An FC pathis made up of the server’s host port ID, the storage subsystem port WWN(worldwide name), and the SD identifier (ID). The following illustration showsa complete path from the server to each of the SDs on the storagesubsystem:


You can display information about the FC paths on the server/cluster throughthe command line interface (CLI), using the fc-host-port-load, fc-target-port-load, and the sdpath commands.

Load balancing and failure recovery

Load balancing on a storage server is a matter of balancing the loads to thesystem drives (SDs) on the storage subsystems to which the storage serveris connected. LUNs and SDs are a logical division of a group of the physicaldisks of the storage subsystem, and LUNs that are visible to the storageserver are known as SDs, which are and the SD is the basic storage unit ofthe storage subsystem.

The server routes FC traffic to individual SDs over a single FC path,distributing the load across two FC switches and, when possible, across dualactive/active or multi-port RAID controllers.

Following the failure of a preferred path, disk I/O is redistributed amongother (non-optimal) paths. When the server detects reactivation of thepreferred FC path, it once again redistributes disk I/O to use the preferred FCpath.

Default load balancing (load balancing automatically performed by thestorage server) is performed based on the following criteria:


• “Load” is defined as the number of open SDs, regardless of the level of I/Oon each SD. SDs count towards load at the target if they are open to atleast one cluster node; the number of nodes (normally all nodes in acluster, after boot) is not considered.

• Balancing load on RAID controller target ports takes precedence overbalancing load on server FC host ports.

• Balancing load among a subsystem’s RAID controllers takes precedenceover balancing among ports on those controllers.

• In a cluster, choice of RAID controller target port is coordinated betweencluster nodes, so that I/O requests for a given SD do not simultaneouslygo to multiple target ports on the same RAID controller.

You can manually configure load distribution from the CLI (overriding thedefault load balancing performed by the server), using the sdpath command.When manually configuring load balancing using the using the sdpathcommand:• You can configure a preferred server host port and/or a RAID controller

target port for an SD. If both are set, the RAID controller target portpreference takes precedence over the server host port preference. When aspecified port preference cannot be satisfied, port selection falls back toautomatic selection.

• For the SDs visible on the same target port of a RAID controller, youshould either set a preferred RAID controller target port for all SDs or fornone of the SDs. Setting the preferred RAID controller target port foronly some of the SDs visible on any given RAID controller target port maycreate a situation where load distribution is suboptimal.

Note: Manually setting a preferred path is not necessary or recommended.

The sdpath command can also be used to query the current FC path beingused to communicate with each SD. For more information on the sdpathcommand, enter man sdpath command.

To see information about the preferred path, navigate to Home > StorageManagement > System Drives, then select the SD and click details to displaythe System Drive Details page. If available, the FC Path section providesinformation about the path, port, and controller.

Fibre channel statistics

The server provides per-port and overall statistics, in real time, at 10-secondintervals. Historical statistics cover the period since previous server start orstatistics reset. The Fibre Channel Statistics page of the NAS Managerdisplays a histogram showing the number of bytes/second received andtransmitted during the past few minutes.


RAID controllersThe RAID controllers operate as an Active/Active (A/A) pair within the samerack. Both RAID controllers can actively process disk I/O requests. Shouldone of the two RAID controllers fail, the storage server reroutes the I/Otransparently to the other controller, which starts processing disk I/Orequests for both controllers.



2Managing the storage subsystem

NAS Platform storage arrays can be managed using NAS Manager. Commonoperations are:• Changing the rack name, password, or media scan period.• Checking the status of media scan and other operations.• Reviewing events logged by the RAID rack.• Determining the status of physical disks.

□ Supported Hitachi Data Systems storage subsystems

□ Using Hitachi Dynamic Provisioning

□ System drive groups and dynamic write balancing

□ Read balancing utility considerations

□ Snapshots and the file system data redistribution utility

24 Managing the storage subsystemHitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

Supported Hitachi Data Systems storage subsystemsAll Series 3000 and Series 4000 NAS storage servers support storage arraysmanufactured by Hitachi Data Systems Corporation. Supported storagearrays are dependent on server series and model:

ServerSeries Server Model Current Offerings Discontinued, but still

supported

4000 4040, 4060, 4080and 4100

HUS VM, HUS 110, HUS 130,HUS 150, VSP G200, VSPG400, VSP G600, VSP G800,and VSP G1000

AMS 2100, AMS 2300, AMS2500, VSP, USP V, USP VM

3000 3080 and 3090 HUS VM, HUS 110, HUS 130,HUS 150, and VSP G1000

AMS 2100, AMS 2300, AMS2500, VSP, USP V, USP VM

Note: All currently supported HDS storage subsystems support RAID levels1, 5, 6, and 10.

Using Hitachi Dynamic ProvisioningYou can use Hitachi Dynamic Provisioning (HDP) software to improve yourstorage utilization. The HDP software uses storage-based virtualizationlayered on top of RAID technology (RAID on RAID) to enable virtual LUNs(dynamically provisioned volumes, DP-Vols) to draw space from multiple poolvolumes. This aggregated space widens the storage bottleneck by distributingthe I/O to more disks. The greater distribution insulates the server from therealities of the pool volumes (small capacities of individual disks).

If you are using HDP, see the Hitachi NAS Platform Storage Pool and HDPBest Practices (MK-92HNAS048) for recommendations.

Note: When using a storage subsystem, there are commonly used Host ModeOptions (HMOs) and System Option Modes (SOMs) which should be setcorrectly. Contact Hitachi Data Systems customer support for moreinformation.

HDP high-level processThe following flow chart shows the high-level process for provisioning storagewith HDP:

Managing the storage subsystem 25Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

Figure 1 High-level process for HDP provisioning

Understanding HDP thin provisioningThin provisioning allows storage to be allocated to an application without itactually being physically mapped on the storage system until it is actuallyused. Thin provisioning also decouples the logical provisioning of storage toan application from the physical addition of storage capacity to the storagesystem.

For example, using a thinly provisioned HDP pool (a DP pool) allows you tocreate a NAS server storage pool with a reported capacity of 80TiB, but theNAS server actually allocates only 30TiB of real storage when creating the filesystems in the storage pool. If you later add parity groups or pool volumes tothe DP pool, you can expand the file systems in the storage pool beyond30TiB without expanding the storage pool itself.

The ability to expand file systems instead of the storage pool is advantageousbecause it enhances performance by spreading the storage chunks used toexpand a file system across all the SDs and physical disks in the DP pool,


whereas a storage pool expansion limits performance by restricting theindividual chunks to a small number of SDs and physical disks.

Note: Hitachi Data Systems strongly recommends that you always use thinprovisioning with HDP.

The NAS server reads the real space available in a DP pool. When you createor expand a file system, the NAS server checks for available space, then pre-allocates the space needed for that operation. If the DP pool has too littlefree space for the operation to complete successfully, the NAS server safelyaborts the creation or expansion of the file system.

HDP allocates pages of real disk space as the server writes data. The servercan write anywhere on any DP-Vol, but not everywhere, meaning that youcannot exceed the amount of real disk space provided by the pool volumes.• HDP thin provisioning should be used in all new NAS server storage pools.

The advantage of HDP thin provisioning is that you can expand theavailable space by adding a single parity group to the DP-Vol; you do notneed to add space in large increments to avoid performance problems.Furthermore, because all the DP-Vols are in the same NAS serverstripeset, any chunk in the storage pool has access to many different DP-Vols, and thus plenty of back-end channels and plenty of queue depth.The process is as follows:1. When provisioning a new NAS server storage pool, use just enough

real disk space to meet your immediate needs for performance andcapacity.

2. Place all your parity groups into a single HDP pool, then create DP-Volswhose total capacity roughly meets your expected needs for the next18 months or 2 years.

Note: It doesn't really matter if you over- or under-estimateyour capacity needs, because you can easily expand the storagepool beyond the total capacity of the original DP-Vols by addinganother set of DP-Vols.

3. Create a NAS server storage pool on these DP-Vols, placing all the DP-Vols into a single stripeset.Use enough DP-Vols to provide adequate queue depth in the future,after you have added enough parity groups to match the total capacityof the DP-Vols. Four DP-Vols is the bare minimum, but eight DP-Volswill provide better performance than four, and sixteen DP-Vols will befaster than 8 DP-Vols. In practice, a storage pool usually contains aneven number of DP-Vols, and the capacity of each DP-Vol is 8TiB.


Note: If using the CLI span-create command, list all the SDsin the initial span-create command. Do not run a single span-create command, then a series of span-expand commands.

Note: When using an application to create a storage pool,specify all the available SDs when creating the storage pool; donot create a single storage pool on a subset of the availableSDs, then expand that storage pool onto the rest of theavailable SDs.

If there are more than 32 available DP-Vols, create the minimumpossible number of NAS server stripesets consistent with making allstripesets identical, even if this means creating slightly more orslightly fewer DP-Vols than would otherwise have been created. Forexample, if you initially estimate that, in two years, you will need 508TiB DP-Vols, you should now create 48 DP-Vols and make 2 stripesetsof 24 DP-Vols each.

4. To expand the NAS server storage pool beyond the total capacity ofthe original DP-Vols, simply add another, identical set of DP-Vols (referto the span-expand man page for more information).

Related concepts

• Thin provisioning using the NAS server filesystem-thin command onpage 28

Thin provisioning using the NAS server filesystem-thin commandHDP thin provisioning differs from the NAS server filesystem-thincommand thin provisioning. Fundamentally, HDP thin provisioning is usedprimarily to improve read/write performance for the entire storage pool (andthe file systems in that storage pool), while thin provisioning done with theNAS server filesystem-thin command works only on an individual filesystem, making the server report that the file system is larger than itactually is to NFS and CIFS (SMB) clients. It is common to use both types ofthin provisioning on the same NAS server storage pool.

With thin provisioning provided by the NAS server filesystem-thincommand, the NAS server reports that a file system is larger than the actualcapacity currently allocated to the file system. NFS and CIFS (SMB) clients,and applications will report that the file system is actually larger than theactual amount of disk storage capacity currently allocated to the file system.Typically, this type of this provisioning is used so that the purchase of actualdisk capacity can be delayed, or so that an application that 'complains' abouta lack of available disk space can be run on a file system that can auto-expand as needed.


The process is as follows:1. Create a small file system.2. Confine it (using the filesystem-confine command) to a larger

capacity.3. Use the filesystem-thin command to enable thin provisioning on that

file system.

Now, if applications or NFS and CIFS clients ask for the capacity of the filesystem, they are told that the capacity is the confined amount, even thoughthe actual amount of disk capacity allocated to the file system is less. As datagets written to the file system, the file system will grow, but only up to themaximum amount to which the file system has been confined, and no further.

Note: A NAS server storage pool can contain a mixture of thinly and thicklyprovisioned file systems.

Related concepts

• Understanding HDP thin provisioning on page 26

Deciding how far to over-provision storageWhen using HDP, you must over-commit the storage for a DP pool to areasonable point. This section can help you decide what makes sense of yoursituation.

The total capacity of the DP-Vols should exceed the total capacity of theparity groups or pool volumes by a factor of 2:1 or 3:1, depending on howfar you expect the storage pool to expand. The total capacity of the DP-Volscreated when the storage pool was initially set up does not constrain theeventual size of the storage pool.

For example, if you have 20TiB of storage and the storage pool may need toexpand to 50TiB later on, you should set up 50TB of DP-Vols. If you everneed to grow the storage pool beyond 50TiB, you can add further DP-Vols.

Limits on thin provisioning:• You can make the storage pool capacity larger than the total capacity of

the DP-Vols that you created at the outset by adding more DP-Vols later.• For HDP, the storage requires an amount of memory that is proportional to

the capacity of the large, virtual DP-Vols, rather than to the smaller, realparity groups or pool volumes. Therefore, consider the following:○ Massive over-commitment causes storage to run out of memory

prematurely.○ Enterprise storage uses separate boards called shared memory.

Related tasks

• Configuring storage to use HDP on page 31


Understanding how HDP works with HNASUsing HDP with HNAS provides many benefits.

HDP with HNAS provides the following benefits:• Improves performance by striping I/O across all available disks• Supports scalability of larger LUs (typically up to 64TiB)• Reduces the need to use the span-expand command, and eliminates

dynamic read balancing (DRB) limitations. When HDP thin provisioning isused, a pool can be expanded in small increments any number of times.However, if you expand a storage pool, make the increments as large asthe initial size of the storage pool to avoid performance problems.

• File system creation or expansion still fails safely, even in the presence ofthinly provisioned pools

To fully realize those benefits, see the HDP configuration guidelines in theHitachi NAS Platform Storage Pool and HDP Best Practices (MK-92HNAS048).

As a general rule, you should make a rough forecast of how much datastorage capacity will be needed in the next 12 to 24 months, then configureyour DP-Vols to be just a little larger than your estimate. If youoverestimated your data storage requirements, not too much space will havebeen wasted; if you underestimated your data storage requirements, you canalways add a second, equally large stripeset using the span-expandcommand, then continue to expand the DP pool in increments as small asdesired.

Some limitations with HDP thin provisioning and HNAS exist. Consider thefollowing:• Some storage arrays and systems do not over-commit by more than a

factor of ten to one.• The amount of memory the storage needs for HDP is proportional to the

size of the (large, virtual) DP-Vols, not the (smaller, real) pool volumes.Therefore, massive over-commitment causes the storage to prematurelyrun out of memory.

Configuring storage for HDP and HNASYou must configure the storage so the HDP software and the HNAS systemcan work together.

Procedure

1. Make every new HDP pool thinly provisioned.2. Create enough DP-Vols to meet the expected size of the span and

provide enough queue depth. For more information on queue depth, seethe sd-queue-depth man page.

3. Wait for the pool to finish formatting.


If the formatting has not completed, the pool may be so slow that itappears to the server that the HDP pool has failed.

Related concepts

• Configuration guidelines for HNAS with HDP on page 33

Configuring storage to use HDPWhen you configure storage to work with HDP, follow this section. You willalso need to consult the HDP software documentation.

Procedure

1. Place several real LUs (pool volumes) in a DP Pool.2. Create several virtual LUs (DP-Vols) on the HDP pool with storage

configurator software).Perform this step using the Hitachi storage configuration application foryour storage subsystem.

3. Give the DP-Vols host LUNs, so the server recognizes them as SDs.To convert the DP-Vols host LUNs to the device identifiers required by thespan-create command:• If you are not using HDP, use the command sd-list --raid-names

--sort kd to display a table listing the SD, the SD identifier, and theSD's LUN.

• If you are using HDP, use the command sd-list --hdp to display atable listing the SD labels and the HDP pool number.

Related concepts

• Deciding how far to over-provision storage on page 29

Before deleting DP-VolsThere are some steps you must take before you delete a DP-Vol.

Note: See the CLI man pages for detailed information about commands.

Important: The span-delete --reuse-dp-vols command imposes extrarequirements. Hitachi Data Systems strongly recommends you read the manpage before trying to use this command.

Procedure

1. Delete the file systems on the storage pool (span in the CLI) that coversthe DP-Vols you want to delete.

2. Use the span-delete command to delete the storage pool that use theDP-Vol you plan to delete.

3. If you plan to reuse the DP-Vols, use span-delete --reuse-dp-vols toavoid space-leakage.


Disable zero page reclaimHDP offers the ability to unmap individual pages within a file system. Thiscapability is called zero page reclaim (ZPR).

Consult the Hitachi Dynamic Provisioning software documentation for moreinformation about ZPR.

Important: ZPR must be turned off when HDP is used with HNAS servers.

Procedure

Confirm that ZPR is turned off when you are using HDP with HNASservers.

Configuration considerations when using HDP with HNASWhen your system must work with HDP-based storage, there are things toremember that affect both the storage and the HNAS server. For detailedinformation on using HDP with your particular storage, consult the HDPsoftware documentation.

When using HDP pools with HNAS servers, consider the following:• All current Hitachi Data Systems storage supports HDP as a licensed

option.• All HNAS servers support HDP without requiring a server-side license.• The HDP software has no effect on protocol clients, such as NFS, and CIFS.• Each DP-Vol draws space from multiple pool volumes, which helps to

relieve the storage bottleneck by distributing I/O to more disks.• If you attempt to create or expand a file system and either there are not

enough free chunks on the span or there is not enough free disk space onthe DP pool, the server will make space by recycling one or more deletedfile systems.

Note: When you recycle or delete a file system, the amount of freespace shown in a storage configurator such as Hitachi CommandSuite, Hitachi Storage Advisor, or Hitachi Storage Navigator, doesnot reflect the new space until after you have run the span-unmap-vacated-chunks command. Do not run this command unnecessarily,since it has a severe performance impact.

• Recycling a file system causes the chunks that stored the file system datato be moved into the vacated-chunks-list, which contains records of whichfreed chunks were used by which file system.

• Creating or expanding a file system draws space from the vacated-chunks-list, if any is available, without using new space from the HDP pool. Anyfurther space is pre-allocated at once.


• Writing to a file system costs no space because that space was pre-allocated.

• When using HDP pools and you create or expand a storage pool, you mustuse the DP-Vols from a single DP pool. This rule applies whether you areusing the CLI or the NAS Manager. Later storage pool expansions can bedone using storage from storage from different DP pools.

Using HDP pools with an HNAS serverYou must configure the HNAS system so that the HNAS software and HDPsoftware can work together.

Important: No configuration changes are required on the HNAS to work withHDP pools.

In general, the process to use HDP pools with an HNAS server is as follows:

Procedure

1. Configure the HDP pools and DP-Vols on the storage:a. Create the HDP pools on the storage.b. Create the DP-Vols to be used in the NAS server storage pool (span).

2. Use the HDP-based storage on the HNAS server:a. Allow access to the DP-Vols (SDs) on the NAS server. You can allow

access using Web Manager of the command line interface.b. Using the HDP DP-Vols, create the HNAS server storage pool.

Note: If you plan on creating tiered file systems, you mustcreate a tiered storage pool using DP-Vols from two HDP pools.

c. Create the file system on the storage pool.d. Format the file system.e. Mount the file system.If you are using the command line interface, the filesystem-create -b4 or filesystem-create -b32 syntax can create, format, and mount afile system in a single step. See the Command Line Reference for yoursystem.

Related concepts


Configuration guidelines for HNAS with HDPFollow these guidelines for best results.

Hitachi Data Systems recommends the following configurations:

On the storage:

• Make every new HDP pool thinly provisioned.


• Create enough DP-Vols to meet the expected size of the storage pool andto provide enough queue depth (minimum of four).

• Limit each HDP pool to hosting only a single NAS server storage pool. Withthe exception of tiered file systems, if you need fifty NAS server storagepools, create fifty HDP pools.

• Do not share an HDP pool between two or more clusters.• Do not share an HDP pool between an HNAS system and a foreign server.

On the NAS server:

• To avoid server timeouts when creating a new NAS server storage pool/span, wait for the HDP pool to finish formatting before creating the NASserver storage pool/span using NAS Manager or the span-createcommand.

Note: If the HDP pool has finished formatting, but the NAS serverdoes not detect the new DP-Vols, run the scsi-refresh CLIcommand so the NAS server will detect the new DP-Vols.

• For tiered file systems, you must use two HDP pools when creating theNAS server storage pool that will host the tiered file systems.

• Do not mix HDP DP-Vols and plain parity groups in a single NAS serverstorage pool/span. However, it is acceptable for some NAS server storagepools/spans to use HDP DP-Vols while other NAS server storage pools useparity groups.

• For best performance, when creating a new NAS server storage pool basedon HDP DP-Vols, specify all the DP-Vols in a single span-create commandor NAS Manager equivalent. Do not create the storage pool on just a fewof the DP-Vols and then make a series of small expansions.

• Create as many file systems as needed and expand them as often asnecessary (or allow them to auto-expand). For maximum flexibility andresponsiveness, create small file systems and allow them to auto-expandas data is written to them.

Note: The maximum size of a newly created file system on an HDPpool is 1TiB, but file system expansions can increase the file systemsize to the maximum limit supported by your NAS server andunderlying storage.

Related tasks

• Using HDP pools with an HNAS server on page 33• Configuring storage for HDP and HNAS on page 30• Creating HDP pools with tiered storage on page 37

Configuring storage pool pre-allocation


In normal operation, on HDP, the NAS Server prevents you from running outof real disk space. If creating or expanding a filesystem would use morespace than is available on the parity groups or pool volumes in the DP pool,the operation fails safely. Without this check, if the DP pool ran out of space,write operations would fail and file systems would be forcibly unmounted.

An important part of this strategy is to pre-allocate real disk space for all theHDP pages in a chunk when that chunk is allocated to a file system. Theserver achieves this pre-allocation by writing one non-zero block to each HDPpage. Without this pre-allocation, the free space on the DP pool would fallsome time later, when data was first written to the HDP pages in the chunk;meanwhile, the server would overestimate the amount of free space on theDP pool, and would be in danger of running out of space.

Although pre-allocation protects your system from running out of disk spaceon a DP pool, it has a number of disadvantages:

• It slows down filesystem creation and expansion.• Owing to the amount of time that it takes, the system limits the size of a

new filesystem or a single manual expansion.• Pre-allocation writes places stress on the storage, and flushes data out of

the cache.• Pre-allocation is not a sufficient safeguard if your storage system uses

FMDs that perform data-compression. On such systems, deletingcompressible data from a file system and writing incompressible data in itsplace can cause the system to run out of space. The Administrator istherefore responsible for monitoring free space, expanding filesystems andadding physical media as required.

It is possible to disable pre-allocation and enable the Administrator tomanage free space, by using the span-hdp-preallocation command. Forfurther information, see the span-hdp-preallocation man page in theCommand Line Reference.

Caution: When using this command, it is essential to ensure that no DP poolever runs out of space. Otherwise, the file systems would be forciblyunmounted and would be unable to remount.

You can re-enable pre-allocation at any time using the same command.However, because some chunks may recently have been allocated without thepre-allocations taking place, some HDP pages in the file systems may not yetbe mapped to real storage, and the server may overestimate the amount offree space available on one or more DP pools. It is therefore necessary toprevent filesystem auto-expansion until the server has written to all of theHDP pages in allocated chunks and obtained an accurate assessment of theamount of free space on DP pools.

Related tasks

• Creating a storage pool using the CLI on page 50


Performing large file system expansionsLarge file system expansions should not be performed when the system isheavily loaded.

Note: The following warning against large expansions applies only when youhaven't used span-hdp-preallocation to disable pre-alloc writes.

The system is heavily loaded when:• experiencing heavy I/O load, especially write load• rebuilding a parity group or pool volume after a disk failure• formatting new disk space• zero-initializing space after recent use of the span-unmap-vacated-chunks

command.

Therefore, if a large expansion is requested, the filesystem-expandcommand prompts you to supply the --storage-is-lightly-loaded switchand will not proceed without it.

Upgrading from older HNAS systemsAny pre-existing storage pool (span in the CLI) should be left thicklyprovisioned after a recent upgrade.

Note: Run span-unmap-vacated-chunks --exhaustive to reclaim spacefrom any deleted file systems and wait for the zero initialization to finish.Thin provisioning can safely be used once the total space used on the poolequals the total size of the file systems on the storage pool.

When upgrading from an older version of an HNAS system, be aware thatcertain conditions produce the results and restrictions described here.

The conditions:• A storage pool (span in the CLI) that was created in an earlier release

violates the SDs in each stripeset must come from one HDP poolrestriction.

• A storage pool (span) contains a mixture of HDP DP-Vols and plain paritygroups. This configuration is unsupported.

The following results are produced:

• Events will be logged at boot time.• The span-list and trouble span will issue warnings.• Some operations will fail cleanly, for example:

○ Cannot create a file system.○ Cannot expand a file system.○ Cannot delete a file system.

• You can still load Cod.• You can still mount file systems.


Note: span-hdp-thickly-provisioned will waive the usual rules if you willnever use thin provisioning on the DP pool that hosts the span.

Using HDP storageWhen working with HNAS systems, the HDP software supports up to twolevels of tiered storage (Tier 0 and Tier 1).

See the Hitachi NAS Platform Storage Pool and HDP Best Practices(MK-92HNAS048) for recommendations.

Considerations when using HDP poolsConsider the following when using the HDP pools:

As with storage pools based on parity groups:• Deleting a file system is not always permanent. Sometimes files systems

are recoverable from the recycle bin or by using the filesystem-undeletecommand.

• Recycling a file system is permanent.

Unlike storage pools based on parity groups, on HDP-based storage pools:• Freed chunks move to the vacated-chunks-list, which is stored in Cod.• Vacated chunks are reused when you create or expand file systems the

same storage pool.• By reusing the same chunks, the server avoids exhausting space

prematurely. Reusing chunks from recycled file systems allows the serverto avoid wasting real disk space on deleted data. Instead, the serverreallocates chunks from recycled file systems to other file systems.

Creating an HDP pool with untiered storageCreate the HDP pool and DP volumes for NAS server.

With untiered storage, tiers are not used. The metadata and the data resideon the same tier. The server has no real perception of tiering with untieredstorage.

Creating HDP pools with tiered storageMost storage pools reside on a single DP pool.

Important: The HNAS systems support up to two levels of tiered storage(Tier 0 and Tier 1).

With tiered storage, the metadata is kept on Tier 0 and the data on Tier 1.Tier 0 should be smaller than Tier 1, but should consist of faster storage. Ingeneral, the amount of real storage allocated for Tier 0 should be about 10%the size of Tier 1. The metadata Tier 0 it will contain is more compact thanuser data but is accessed more often. A single DP pool can support both themetadata and data tiers.


Related concepts


Creating storage pools with DP pools from HDP storageAfter you have created an HDP pool with tiered or untiered storage, you canuse DP-Vols to create storage pools.

See the CLI man pages for detailed information about commands.

Procedure

1. Use the command span-create or the NAS Manager equivalent to createthe storage pool on the first HDP pool’s DP-Vols.

2. Use the command span-expand to expand the storage pool on to theHDP second pool’s DP-Vols.Expanding the storage pool at the outset avoids the disadvantages ofexpanding it on a mature span. This is the only recommended exceptionto the rule of one pool per storage pool and one storage pool per pool.

3. When necessary, add new pool volumes to whichever pool needs them.Use the following steps:

Note: Do not use dynamic read balancing (DRB, the command fs-read-balancer) at this step.

a. Add parity groups (PGs) or pool volumes.b. If the amount of storage in the affected pool exceeds the total size of

its DP-Vols, add more DP-Vols and use span-expand.

Moving free space between storage poolsYou can move free space between storage pools, but you should firstthoughtfully consider the implications because of the strong performanceimpacts.

The span-unmap-vacated-chunks launches a background thread that mayrun for seconds, minutes, hours, days or even months, and which can bemonitored and managing using commands mentioned in its man page.

The free space on the DP pool will keep increasing as long as this backgroundthread runs.

On configurations where the storage has to zero-initialize (overwrite withzeros) HDP pages before they can be reused, the free space on the pool maywell continue to increase even after the unmapping thread terminates.

The performance of all DP pools on the affected array will be lower than usualuntil free space has finished increasing, but DP pools on other storage arrayswill be unaffected.


Related tasks

• Deleting a storage pool on page 62

Unmapper use and why to avoid itThe Hitachi Data Systems recommended best practice is to dedicate eachpool to a single storage pool (span on the CLI). However, although notrecommended, should a situation arise where multiple storage pools (spans)exist on a single pool, you can use the unmapper feature to move spacebetween the storage pools on that pool.

Important: Using the unmapper commands can have serious consequences.Hitachi Data Systems strongly recommends that you read the CLI man pagesfor each command.

Considerations:• Unmapping vacated chunks does free space, but the performance impact

is extreme. Never unmap chunks just to affect the appearance of availablestorage.

• You cannot boot HNAS version 12.0 or earlier into the cluster while anystorage pool (span) has a non-empty vacated chunks list. Should you needto downgrade to 12.0 or earlier, use the span-vacated-chunks commandto identify storage pools whose vacated-chunks-lists are not empty. Then,use the span-unmap-vacated-chunks command on each of those storagepools. Finally, wait for events to indicate that unmapping has finished oneach storage pool. There is no need to wait for zero-initialization(overwriting to zeros) to take place inside the storage.

• You can unmap space on any number of spans at one time, butperformance is further impacted.

• The server has no commands for monitoring or managing the HDP zero-init process. Once the process starts, you have to wait until it finishes. Thetime can exceed many hours, even weeks in some cases.

Further reasons to avoid using the unmapper:• The span-unmap-vacated-chunks command launches a background

process that takes a very long time to run.• On most storage configurations, an HDP page cannot be reused

immediately after being unmapped. For security reasons, the page mustfirst be zero-initialized to overwrite the previous page with zeros. Thisprocess occurs inside the storage, and it cannot be monitored or managedby commands on the server.

• Until pages have been zeroed, they’re not available for reuse.• Zero-initialize can impact the performance of the connected storage and

also that of other HDP pools.

The mapper feature uses the following commands:• span-vacated-chunks displays the number of vacated chunks in a storage

pool and the progress of any unmapper.


• span-stop-unmapping cancels an unmapper without losing completedwork.

• span-throttle-unmapping helps you avoid long queues of pages waitingto be zero-initialized.

The only tested way to minimize the unmapper performance impact is tochange the format priority from Normal to Host Access. Doing so makesformatting slower but enables the server to keep running.

Using the unmapper to move space between storage poolsIf, after thoughtfully considering the consequences associated with use of theunmapper, you decide it is worth the significant performance impact, you canuse the following steps to move space between storage pools.

Unmapping chunks (using the unmapper) does not increase the amount offree space in a storage pool; instead, the unmapper:1. Takes chunks out of the VC list and returns the underlying HDP pages to

the DP pool.2. Removes the capacity from the chunks listed in the vacated chunks list

by unmapping the space.3. Zero-initializes the space.4. Returns the freed capacity to the underlying DP pool.

After being added to the DP pool, the freed space can be used by otherstorage pools.

Note: See the CLI man pages, or the Command Line Reference for detailedinformation about the commands mentioned below.

The following procedure describes how to move space from storage pool S tostorage pool T. This procedure is based on a configuration where both storagepools are based on DP-Vols from the same DP pool, and that DP pool is thinlyprovisioned.

Procedure

1. Delete and recycle a file system from storage pool S (Span S).2. Run span-unmap-vacated-chunks on storage pool S.3. Run the span-list --sds T command, and look at the amount of free

space in the DP pool.When the output from the span-list --sds T command shows that theDP pool has enough free space, create a new file system in storage poolT and/or expand one or more file systems in storage pool T.

Note: It may take a significant amount of time for the zero-initialization process inside the storage to complete and theamount of space free in the DP pool to increase. The amount of


time it takes for this process is dependent on the type of storage,the amount of space being initialized, and utilization of the storagesubsystem.

If it takes too long to add the freed space to the DP pool, you can expandstorage pool T onto a different DP pool (one that has available space).

System drive groups and dynamic write balancingWhen used with a storage pool based on parity groups (RAID groups),dynamic write balancing (DWB) maximizes performance by ensuring that theNAS server writes to as many SDs as possible. Dynamic write balancing alsoimproves flexibility by letting the server reflect the physical characteristics ofthe storage without the need to reconfigure spans.

Dynamic write balancing is enabled by default for storage pool based onparity groups (RAID groups).

In previous releases, during a write operation, the writing the NAS servercould write to a single stripeset at any time. The stripeset being written tomay contain only a small fraction of all the SDs in the storage pool. Thisproduced three performance problems during write operations:1. A storage bottleneck is created because all writes are going to a single

stripeset, regardless of the number of SDs in the storage pool.2. If the stripesets vary in performance (for example, some SDs may be on

higher performance storage or may contain more SDs), the writeperformance of the file system will vary over time, depending on thestripeset characteristics.

3. If more storage is added to the storage pool, the file system's writeperformance does not immediately improve; it will improve only afternew chunks have been added to the file system. However, writeperformance will fall again when writing to older chunks.

Dynamic Write Balancing (DWB) solves these problems by writing to all SDsin parallel.

To implement Dynamic Write Balancing, the NAS server requires someknowledge of the physical configuration of the storage. SDs must be assignedto SD groups, with each SD group typically corresponding to one RAID group.After SD groups have been configured, write operations are associated withSD groups rather than with SDs; within each SD group, the NAS server willscan the whole of one SD for free space before moving on to the next SD. Formore information on SD Groups, see the sd-group man page.

Note: System drive groups are not used with storage pools based ondynamically provisioned volumes (DP-Vols) created with Hitachi DynamicProvisioning (HDP) software.


Optimizing dynamic write balancing performance

Although dynamic write balancing removes many of the restrictions ofprevious allocation schemes, a few important guidelines still apply:• Hitachi NAS Platform only: Divide each RAID array into as few SDs as

possible, and use multiples of four (4) SDs whenever possible. Forexample: 4-8-12.

• For Unified NAS Module only: Divide each RAID array into as few SDs aspossible, using even numbers of SDs. For example: 4-6-8 in a stripe set.

Note: Some software user interfaces have their own algorithms,usually creating even numbers of SDs in a stripe set

• Never divide storage into dozens of tiny SDs, and then create a storagepool from the many small SDs.

All the SDs in a RAID group or an HDP pool should be used in the samestorage pool. If multiple SDs in an SD group are shared between storagepools, the SD group mechanism will not prevent the server from writing toseveral of the shared SDs at once. Writing to several of the shared SDs atonce can cause a performance reduction, because one HDP pool may bedriving the storage pool very hard, causing a slow-down on the other storagepools using the same resources.

To help with server to storage troubleshooting, keep the following in mind:1. The server logs an event when a system drive becomes degraded on an

HUS 1x0 array. For example:Warning: Device 0 (span "SPAN" ID 5B95587A30A2A328) : Devicereporting : SD 0: SCSI Lan Sense LU status reports DEGRADED"

2. A trouble command reporter will identify an SD that may have higherthan average response times.

3. The output of the scsi-devices command includes the internal LUNvalue of any SD.

4. For solutions with Hitachi Universal Replicator and TrueCopy, if theprimary SDs are visible only to one server (or cluster) and thesecondaries only to another server (or cluster), the sd-mirror-remotelycommand must be used add SD mirror relationship information to theserver's the internal database.

5. Make sure that the Command Queue Expansion Mode is enabled on HUS1x0 arrays. Enabling the Command Queue Expansion Mode prevents theNAS server from inadvertently overloading the HUS ports with excessiveamounts of IO.


Read balancing utility considerationsThe read balancing utility moves data onto newly allocated chunks on a newstripeset of a storage pool. The read balancing utility is run after you haveadded a new stripeset to a storage pool and then expanded a file systemonto the new stripeset.

When used with a storage pool based on parity groups (RAID groups), readbalancing helps to redistribute static datasets. After adding SDs to a span,the file system data redistribution utility causes data to be re-written to anew location, which will be the least utilized SD groups (the most recentlyadded SDs) resulting in more balanced utilization of SD groups.

Note: Read balancing is not used with storage pools based on dynamicallyprovisioned volumes (DP-Vols) created with Hitachi Dynamic Provisioning(HDP) software. The read balancing utility is not needed when HDP is used,because HDP automatically moves data to new disks without requiring filesystem expansion.

The file system data redistribution utility should be run immediately afterexpanding a file system, to rewrite data from already-used storage chunks tochunks on the newly added stripeset. The file system data redistributionutility may be run only once per file system expansion. If you run the dataredistribution utility more than once, or after an application has written asignificant amount of data into the expanded file system, the utility will eitherrefuse to run or will produce unpredictable results.

For the utility to be run effectively, file system should under-utilize the SDs ofthe most recently added stripeset (the closer to 0%, the better) and the filesystem should over-utilize the SDs of the older stripesets (the closer to100%, the better). Use the command span-space-distribution to obtainthis information.

Note: Each time a stripeset is added, you must expand file system and thenrun the file system data redistribution utility. In other words, you cannot addthree new stripesets, expand the file system to use chunks from all threestripesets, and then run the utility. When adding more than one stripeset, forevery stripeset you add, you must:1. Add the stripeset.2. Expand all file systems in the storage pool by the same proportion as you

expanded the storage pool.For example, if you double the capacity of a storage pool, the size of allfile systems in that storage pool should also be doubled. If you expandthe storage pool capacity by 33% then the file systems in that storagepool should also be expanded by 33%.

3. Run the file system data redistribution utility.


Some use cases for using the read balancing utility after adding SDs to astorage pool would be:• The customer is expecting to double the amount of data in the file system,

but access to the existing data is largely read-only.Immediately doubling the file system size and re-balancing would makesense, because then the file system’s free space should be distributedroughly equally across both the original SDs and the new SDs of thestorage pool. In this case, re-balancing allows the file system to use all thestorage devices in all the SDs as the data grows. If the file system size isincreased little by little, the media of the new SDs of the storage pool willbe used as the file system expands.

• The customer is not expecting the amount data to grow, it is largely static,and the current SDs are a bottleneck in the READ path. Doubling the filesystem size and re-balancing should move half the READ load onto thenew SDs.

The file system data redistribution utility is designed to operate when a filesystem is expanded into new storage after SDs have been added to a storagepool when the file system is nearly full. However, storage may also be addedto a storage pool for other reasons:• To increase performance.• To prevent the file system from becoming completely full.

To achieve the desired results in either of these situations:1. Add the stripeset.2. Issue the CLI command filesystem-expand (--by <GiB> | --to

<GiB>) --on-stripeset X <filesystem-instance-name> command,where X is the number of the stripeset you just added (note thatstripeset numbering begins with 0, so if your storage pool has threestripesets, the newest is stripeset number 2).

3. Run the file system data redistribution utility.4. Repeat steps 2 and 3 for each file system in the storage pool.

Snapshots and the file system data redistribution utilityWhen the file system data redistribution utility is run and file systemsnapshots are enabled, the old data is preserved, because the dataredistribution utility cannot balance snapshot allocations. As a result, filesystem snapshot space usage will grow, consuming a lot of disk space. Thespace used by these snapshots is not freed until all snapshots present whenthe file system data redistribution utility was started have been deleted.

There are four options available to recover the space used by snapshots:1. Allow the snapshots to be deleted according to the snapshot

configuration.


This is the slowest option for recovering space, but it can be used inscenarios where the space won’t be required immediately after the filesystem data redistribution utility completes.

2. Manually delete snapshots after running the file system dataredistribution utility.This option recovers space more quickly than option 1.

3. Manually kill snapshots after running the file system data redistributionutility.This option also recovers space more quickly than options 1 or 2, but itrequires that the file system be taken offline.

4. Disable snapshots (and therefore backups) and kill/delete existentsnapshots before running the file system data redistribution utility.This option avoids the snapshot space usage problem altogether.


3Using a storage pool

Storage pools contain one or more file systems, which consume space fromthe storage pool upon creation or expansion. A storage pool can also be usedto control the auto-expansion policy for all of the file systems created in thestorage pool. The following procedures describe how to create, delete,expand, remove from service, and rename a storage pool.

Once access is allowed to one system drive (SD) in a storage pool, thatstorage pool becomes visible in NAS Manager. If access is denied to all SDs ina storage pool, the storage pool is not visible in NAS Manager.

□ Creating storage pools

□ Adding the metadata tier

□ Expanding storage pools

□ Configuring automatic file system expansion for an entire storage pool

□ Renaming a storage pool

□ Reducing the size of a storage pool

□ Denying access to a storage pool

□ Allowing access to a storage pool

□ Deleting a storage pool

46 Using a storage poolHitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

Creating storage poolsYou can create storage pools from either the GUI or CLI.

Creating a storage pool using the GUIWith available SDs, administrators can create a storage pool at any time.After being created, a storage pool can be expanded until it contains up to256 SDs.

When creating a tiered storage pool, to attain optimal performance, makesure that the SDs of the metadata tier (Tier 0) are on the highestperformance storage type.

After the storage pool has been created, smaller file systems can be createdin the pool for more granular storage provisioning.

Procedure

1. Navigate to Home > Storage Management > Storage Pools, andclick create to launch the Storage Pool Wizard.

2. Select either Untiered Storage Pool or Tiered Storage Pool.

Note: If you are creating a tiered storage pool, this will createTier 1 User-data, of the tiered storage pool).

3. From the list of available SDs, select the SDs for the storage pool/tier.

Select at least two SDs for use in building the new storage pool/tier. Toselect an SD, fill the check box next to the ID (Label).

An untiered storage pool cannot contain SDs on RAID arrays withdifferent manufacturers, disk types, or RAID levels. Any attempt tocreate a storage pool from such dissimilar SDs will be refused.

Using a storage pool 47Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

A tiered storage pool can contain SDs on RAID arrays with differentmanufacturers, or disk types, as long as they are in different tiers. Atiered storage pool cannot, however, contain SDs with different RAIDlevels. Any attempt to create a storage pool with SDs that have differentRAID levels will be refused.

For the most efficient use of storage capacity in an untiered storage poolor in a tier of a tiered storage pool, Hitachi Data Systems customersupport strongly recommends that all SDs be of the same capacity,width, and stripe size, and disk size; however, after first acknowledging awarning prompt, you can create a storage pool with SDs that are notidentically configured.

4. Specify the storage pool label.• Storage pool labels are not case sensitive, but they do preserve case

(labels will be kept as entered, in any combination of upper and lowercase characters). Also, storage pool labels may not contain spaces orany of the following special characters:○ Double quote (")○ Single quote (')○ Ampersand (&)○ Asterisk (*)○ Slash (/)○ Backslash (\)○ Colon (:)○ Semicolon (;)○ Greater than (>)○ Less than (<)○ Question mark (?)○ Vertical bar/pipe (|)

Tip: Guidelines for choosing a good storage pool label include:

• The label should reflect the contents of the storage pool. Reasonablyshort, but distinctive and descriptive labels will help to guard againstmistakes.

• Storage pool labels should be unique across the entire site, not just onthe local cluster. If you move storage between servers or clusters,duplicate names will cause needless difficulty. Also, generic labels suchas 'SAS_POOL_0' are best avoided, because they are not mnemonicand they are more likely to be duplicated among the clusters at a site.See the man page or the Command Line Reference for the storage-based-snapshot-label command for an explanation of rules and thevarious name types and interactions.


• A storage pool label should not be the same as a file system label.• The storage pool label should not resemble a device ID (it should not

be just a sequence of 1-4 digits). 5. Verify your settings, and click next to display a summary page.

The summary page displays the settings that will be used to create thestorage pool/tier.

If you have already set up mirrored SDs for disaster preparedness orreplication purposes, and you want the server to be aware of the mirrorrelationship, fill the Look For Mirrored System Drives check box.

Note: Before filling the Look For Mirrored System Drives checkbox, you must have finished configuring the mirrored SDs usingthe RAID tools appropriate for the array hosting the mirrored SDs.For example, for Hitachi Data Systems Corporation storage arrays,you would use True Copy to create the mirrored SDs.

Note: The Look For Mirrored System Drives checkbox is usedonly when setting up mirrored SD relationships using HitachiTrueCopy. The Look For Mirrored System Drives checkbox isnot used with Hitachi Universal Replicator (HUR), Hitachi Copy-on-Write (CoW), or Hitachi Global-active device (GAD) software.

The guideline chunk size is specified when creating the storage pool.6. After you have reviewed the information, click create to create the

storage pool/tier.7. If you are creating an untiered storage pool, you can now either:

• Click yes to create file systems (refer to the File ServicesAdministration Guide for information on creating file systems).

• Click no to return to the Storage Pools page without creating filesystems.

8. If you are creating a tiered storage pool, you can now either:• Click no to return to the Storage Pools page if you do not want to

create the metadata tier (Tier 0) of a tiered storage pool.• Click yes to display the next page of the wizard, which you use to

create the metadata tier (Tier 0) of a tiered storage pool.1. Specify which SDs to use in the tier by filling the check box next

to the SD label.2. Click next to display the next page of the wizard, which is a

summary page.3. If you have mirrored SDs, for disaster preparedness or replication

purposes, and you want the server to be aware of the mirrorrelationship, fill the Look For Mirrored System Drives checkbox.


4. After you have reviewed the information, click add to create themetadata (Tier 0) tier of the storage pool. A confirmation dialogwill appear, and you can now choose to create file systems in thestorage pool, or you can return to the Storage Pools page.• Click yes to create file systems (refer to the File Services

Administration Guide for information on creating file systems).• Click no to return to the Storage Pools page.

Note: After the storage pool has been created, it can be filled withfile systems. For more information, see the File ServicesAdministration Guide.

Related concepts

• Storage pool chunks on page 16

Creating a storage pool using the CLIYou can use the CLI to create storage pools.

Note: For detailed information about the span-create command, see the CLIman pages.

Procedure

On the HNAS system, use the span-create command to create a storagepool using the SDs from the DP-Vols (on storage). For more informationabout the span-create command, refer to the Command Line Reference.

Note: If you are using HDP:• To avoid server timeouts when creating a new NAS server

storage pool, wait for the HDP pool to finish formatting beforecreating the NAS server storage pool.If the HDP pool has finished formatting, but the NAS serverdoes not detect the new DP-Vols, run the scsi-refreshcommand so the NAS server will detect the new DP-Vols.

• If you are using HDP thin provisioning, list all the SDs in theinitial span-create command. Do not run a single span-createcommand, then a series of span-expand commands.

Related references

• Configuring storage pool pre-allocation on page 34


Adding the metadata tierIf you created a tiered storage pool, but only defined the SDs for the userdata tier (Tier 1), you must now create the metadata tier (Tier 0).

Note: You can convert an untiered storage pool to a tiered storage pool usingthe span-tier command. For more information about this command, refer tothe Command Line Reference.

To add a tier to a storage pool:

Procedure

1. Navigate to Home > Storage Management > Storage Pools.

2. Select the storage pool to which you want to add the tier. Click details todisplay the Storage Pool Details page.

3. Click the Add a Tier link, to display the Storage Pool Wizard page, tocreate the user data tier.

4. Select the SDs to make up the metadata tier.Using the Storage Pool Wizard page, above, select the SDs for thesecond (user data) tier from the list of available SDs on the page. Toselect an SD for the tier, fill the check box next to the SD ID Label in thefirst column. Verify your settings, then click next to display a summarypage.

5. Review and apply settings.The summary page displays the settings that will be used to create thestorage pool/tier.

If you have already created mirrored SDs for disaster preparedness orreplication purposes, and you want the server to be aware of the mirrorrelationship, fill the Look For Mirrored System Drives checkbox.


Note: Before filling the Look For Mirrored System Drives checkbox, you must have finished configuring the mirrored SDs usingthe RAID tools appropriate for the array hosting the mirrored SDs.For example, for Hitachi Data Systems Corporation storage arrays,you would use True Copy to create the mirrored SDs.

Note: The Look For Mirrored System Drives checkbox is usedonly when setting up mirrored SD relationships using HitachiTrueCopy. The Look For Mirrored System Drives checkbox isnot used with Hitachi Universal Replicator (HUR), Hitachi Copy-on-Write (CoW), or Hitachi Global-active device (GAD) software.

Once you have reviewed the information, click add to create the secondtier of the storage pool.

Note: After the storage pool has been created, it can be filled withfile systems.

6. Complete the creation of the storage pool or tier.After clicking add (in the last step), you will see a confirmation dialog.

You can now click yes to create a file system, or click no to return to theStorage Pools page. If you click yes to create a file system, the CreateFile System page will appear.

Expanding storage poolsTo expand a storage pool, simply add SDs to the storage pool. If the storagepool is tiered, you can add SDs to one tier without adding SDs to the othertier. span-list--sds shows how full each tier is.

If you are using Hitachi Dynamic Provisioning, see the Hitachi NAS PlatformStorage Pool and HDP Best Practices (MK-92HNAS048) for recommendations.

Related references

• Configuring storage pool pre-allocation on page 34

Why use HDP to expand DP-VolsExpanding DP-Vols without using HDP does not have the benefits HDPprovides.

Using HDP to add space provides the following benefits:• You can add disks in small increments, even just a single pool volume.• Data gets restriped.


• Span gets faster and performance remains almost even.

Consider the following use case for using HDP for expanding DP-Vols:

If you originally created a pool containing 10TiB of real storage and eight DP-Vols of 2.5TiB each, totalling 20TiB, the pool is over-committed by a ratio of2:1. As always, a storage pool (span on the CLI) resides on the DP-Vols. Astime goes by, you make a series of expansions of 4TiB each by adding newparity groups or pool volumes. The first expansion increases the amount ofreal storage in the pool to 14TiB and the second expansion takes it to 18TiB.

After each of these expansions, no further action is necessary. However, aftera third 4TiB expansion, the pool contains 22TiB of real storage, but its DP-Vols total only 20TiB. As a result, 2TiB of the disk space that you haveinstalled are inaccessible to the server.

More DP-Vols are needed, but any expansion should always add at least asmany DP-Vols as were provided when the span was created. You musttherefore create a further eight DP-Vols, preferably of the same 2.5TiB as theoriginal ones, and add them to the storage pool by using the span-expandcommand or NAS Manager equivalent. This addition brings the total DP-Volcapacity to 40TiB. No further DP-Vols will be necessary until and unless thereal disk space in the pool is expanded beyond 40TiB.

Expanding space in a thinly provisioned HDP storage poolYou can easily add space to a storage pool that uses thin-provisioned HDP.

The pool formatting process is non-disruptive, so the file systems staymounted during the process.

Note: For detailed information about specific commands and how they areused, see the CLI man pages or the Command Line Reference.

Procedure

1. Create the pool volumes.2. Use the span-confine command to confine the span.3. Add the pool volumes to the HDP pool.

Adding the pool volumes automatically enables the Optimize checkbox.4. Wait for the pool to finish formatting.5. If required, use the span-release command to release the span on the

HNAS system.The HNAS system auto-detects the new space and lets you use it in newor existing file systems.

6. Check that the real disk space in the pool still does not exceed the totalcapacity of the pool's DP-Vols.

Related tasks

• Expanding storage space using DP-Vols on page 54


Expanding storage space using DP-VolsEventually, the total size of the HDP pool volumes reaches the total size ofthe DP-Vols. If the NAS server storage pool (span) needs more space, youcan add space to it.

You can add as many HDP pool volumes as you want; however, you typicallyonly need to add a small amount of space.

Note: See the CLI man pages or the Command Line Reference for detailedinformation about commands.

Procedure

1. Add the new pool volumes to the original pool.2. Add more DP-vols to the same HDP pool.

Note: Make the new DP-Vols the same size and number as youoriginally created. All stripesets must be the same.

3. Wait for formatting to finish.Otherwise, the file systems may auto-expand onto the new storage andfind it so slow that the entire span fails.

4. Use the span-expand command to expand the span on to the new DP-Vols.

Related tasks

• Expanding space in a thinly provisioned HDP storage pool on page 53

Expanding a non-HDP storage pool or tier

Procedure

1. Navigate to Home > Storage Management > Storage Pools todisplay the Storage Pools page.

2. Fill the check box next to the label of the storage pool you want toexpand, and click details.

If the storage pool is an untiered storage pool, the Storage PoolDetails page looks like the following:


To display the available system drives to add to the storage pool, clickexpand. The Storage Pool Wizard page is displayed.

If the storage pool is a tiered storage pool, the Storage Pool Detailspage looks like the following:


3. To display the available system drives to add to a tier, select the tier youwant to expand, and click expand to display the Storage Pool Wizardpage.

4. Fill the check box next to the label of the system drive you want to add,then click next to display the next Storage Pool Wizard page.

5. Click expand to add the SDs to the storage pool/tier.

Configuring automatic file system expansion for an entirestorage pool

Use this procedure to allow or prohibit automatic expansion of all file systemsin the specified storage pool. This setting only affects auto-expansion;manual expansion of file systems in the storage pool is not affected by thissetting.

Procedure

1. Navigate to Home > Storage Management > Storage Pools.


2. Select a storage pool, and click details to display the Storage PoolsDetails page.

3. Configure auto-expansion.You can configure file system auto-expansion at the storage pool level asfollows:• Enable auto-expansion

Even if the storage pool is configured to allow its file systems toautomatically expand, the file systems must also be configured tosupport automatic expansion. After a file system has expanded, itssize cannot be reduced.If file system auto-expansion is currently disabled, you can enable itby clicking enable auto-expansion in the FS Auto-Expansionoption box.

• Disable auto-expansionWhen automatic expansion of a file system is disabled, manualexpansion of file systems in the storage pool is still possible.If file system auto-expansion is currently enabled, you can disable itby clicking disable auto-expansion in the FS Auto-Expansionoption box.


Renaming a storage poolThe name for a storage pool can be changed at any time, without affectingany clients.

Procedure


2. Select a storage pool, and click details.3. Enter a new name in the Label text box, and click rename.

Storage pool namingStorage pool naming rules and conventions.

A storage pool label may be one of the following types:• A base name, which is the name that the storage pool or file system was

given when it was created.All copies of a storage pool have the same base name. All copies of a filesystem have the same base name unless you decide to rename one ormore copies.○ Storage pool labels are not case sensitive, but they do preserve case

(labels will be kept as entered, in any combination of upper and lowercase characters). Also, storage pool labels may not contain spaces orany of the following special characters:– Double quote (")– Single quote (')– Ampersand (&)– Asterisk (*)– Slash (/)– Backslash (\)– Colon (:)– Semicolon (;)


– Greater than (>)– Less than (<)– Question mark (?)– Vertical bar/pipe (|)

Tip: Guidelines for choosing a good storage pool label include:

• The label should reflect the contents of the storage pool. Reasonablyshort, but distinctive and descriptive labels will help to guard againstmistakes.

• Storage pool labels should be unique across the entire site, not just onthe local cluster. If you move storage between servers or clusters,duplicate names will cause needless difficulty. Also, generic labels suchas 'SAS_POOL_0' are best avoided, because they are not mnemonic andthey are more likely to be duplicated among the clusters at a site.See the man page or the Command Line Reference for the storage-based-snapshot-label command for an explanation of rules and thevarious name types and interactions.

• A storage pool label should not be the same as a file system label.• The storage pool label should not resemble a device ID (it should not be

just a sequence of 1-4 digits). • A snapshot name (or snap name), which identifies a single copy of the

data. When you place a storage pool into snapshot mode, and every timeyou add a new snapshot, you specify a new snap name. Every snapshot ofa given storage pool must have a unique snap name, although snapshotsof different storage pools may have the same snapshot name.

Note: A snapshot name follows the same rules about specialcharacters as a storage pool or filesystem base name, but snapname also cannot contain a dash (-).

• An instance name, which is automatically constructed from the base nameand the snapshot name, if snapshots are used. The instance nameidentifies a single copy of a storage pool or file system.For example, if a storage pool labelled 'Accounts' has a snapshot called'Wednesday', the instance name of this snapshot is 'Accounts-Wednesday'.If the storage pool has a file system called 'External' then the instancename of this snapshot of the filesystem is 'External-Wednesday'. Moststorage pool commands and file system commands expect instancenames.

Note: When the server creates or loads a file system, file system details arestored in the server's registry, where they can be displayed by the


filesystem-list-stored command. The server compares these labels asfollows:• Case-insensitively: a storage pool called 'AAA' cannot be loaded at the

same time as a storage pool or file system called 'aaa'.• Objects of the same type. For example, you cannot have two file systems

with the same name, even if they are in different storage pools.• Objects of different types. For example, if you have snapshotted storage

pools with instance names 'Accounts-Main' and 'Accounts-DataMine', youcannot then create a new unsnapshotted storage pool or a file system witha label of Accounts'.

When labelling a storage pool, no storage pool may have the same "basename" or "instance name" as the base or instance name of any other loadedstorage pool or file system.

Reducing the size of a storage poolThe size of a storage pool cannot be reduced.

Denying access to a storage poolTypically, you deny access to a storage pool when you no longer want to usethe storage pool with the local NAS server or cluster. If you want to use thestorage pool with another NAS server or cluster, refer to the Command LineReference page for the span-assign-to-cluster command, or view thespan-assign-to-cluster man page for information on migrating a storagepool safely.

Denying access to a storage pool:

• Unloads the storage pool and its file systems from memory, so that theydo not appear in the NAS Manager list of storage pools or in the results ofthe span-list -f command.

• Unlicenses the host SDs.

• Deletes all the non-volatile memory contents for all file systems in thestorage pool.

Caution: If a file system did not unmount cleanly before access tothe storage pool is denied, all data written by clients andacknowledged by the server, but not yet been written to the storage,will be lost.


• Removes all file system to EVS bindings. If you later allow access to thestorage pool, you will need to assign each file system to an EVS, eitherusing NAS Manager or using the evsfs add command.

Note: Denying access to a storage pool does not delete the storage pool, thefile systems, or the SDs.

Procedure

1. Navigate to Home > Storage Management > File Systems to displaythe File Systems page.

2. Identify the file systems in the storage pool to which you want to denyaccess, then select those file systems.To select a file system, fill the check box next to the file system label.

3. Unmount the selected file systems.Click unmount, and in the confirmation dialog, click OK.

4. Click the Storage Pools shortcut to display a list of all pools, select aparticular storage pool, and click Deny Access; in the confirmationdialog, click OK.

Note: This will also remove the pool from the storage pools list,but it will not be deleted.


Allowing access to a storage poolThis procedure restores access to a storage pool, but can also be used whena storage array previously owned by another server has been physicallyrelocated to be served by another server. The process restores access to theSDs that belong to the storage pool, then restores access to the pool itself.

Note: Before moving the storage pool from one NAS server/cluster toanother, refer to the Command Line Reference page for the span-assign-to-cluster command, or view the span-assign-to-cluster man page forinformation on migrating a storage pool safely.

To allow access to a storage pool:

Procedure

1. Navigate to Home > Storage Management > System Drives.2. Select one of the SDs belonging to the storage pool, and click Allow

Access.3. Select a pool, and click details. In the Details page for that storage

pool, click Allow Access; then, in the Confirmation page, click OK.

Note: To become mountable, each file system in the storage poolmust be associated with an EVS. To do this, navigate to theDetails page for each file system in the storage pool and assign itto an EVS.

Deleting a storage poolA storage pool that does not contain file systems can be deleted at any time.If the storage pool contains file systems, you must unmount and then deletethe file systems before you can delete the storage pool. After the pool hasbeen deleted, its SDs become free and available for use by other storagepools.

Note: For detailed information about specific commands, see the CLI manpages or the Command Line Reference.

If you are using HDP:1. Unmount the file systems in the storage pool using the unmount

command.2. Delete the file systems in the storage pool using the filesystem-delete

--no-undeletion-information command. (If the file systems havealready been deleted, use the span-list-recycle-bin and filesystem-recycle commands to recycle all file systems in the storage pool.)


3. Delete the storage pool. You have two options:• If you do not plan to reuse the DP-Vols, you should:

1. Delete the storage pool using the span-delete --will-delete-dp-vols command.

2. Delete the DP-Vols using the storage configuration application foryour storage system. Optionally, you can then delete the DP poolif it contains no other DP-Vols or storage pools.

• If you do plan to reuse the DP-Vols, you should:1. Return the storage pool's previously-used chunks to the

underlying HDP pool using the span-unmap-vacated-chunkscommand.

2. Monitor the progress of returning the storage pool's previously-used chunks to the underlying HDP pool using the span-vacated-chunks command.When the span-vacated-chunks command no longer reports thatan unmapper is running against the storage pool. There is noneed to wait until the zero-initialisation process inside the storagehas completed.

3. Delete the storage pool using the span-delete --reuse-dp-volscommand. (See the chunk command man page or the CommandLine Reference for detailed information about this command andrelated commands. )

You can now reuse the DP-Vols in another storage pool. If the zero-initialisation process is still running, it may take some time for the fullcapacity of the DP pool to become available.

Note: Running the command span-delete --will-delete-dp-vols, then reusing the DP-Vols in a new storage pool will leakspace and make it difficult to create and expand file systems inthe new storage pool. The difficulty is caused because theserver's list of mapped, reusable chunks is destroyed when youdelete the old storage pool. If you find yourself in this position,use the span-unmap-vacated-chunks --exhaustive commandon the new storage pool and wait for both the unmapping fibreand then the zero-initialisation process inside the storage tocomplete.

If you are not using HDP, use the following procedure.

Procedure



2. Click details for the storage pool you want to delete.The Storage Pool Details page will be displayed.

3. Click delete, then click OK to confirm.

Related concepts

• Moving free space between storage pools on page 38



Hitachi Unified NAS Module and Hitachi NAS Platform Storage Subsystem Administration Guide

Hitachi Data Systems

Corporate Headquarters2845 Lafayette StreetSanta Clara, California 95050-2639U.S.A.www.hds.com

Regional Contact Information

Americas+1 408 970 [email protected]

Europe, Middle East, and Africa+44 (0) 1753 [email protected]

Asia Pacific+852 3189 [email protected]

MK-92HNAS012-09

July 2016

http://www.hds.com

MAILTO:[email protected]



Date post:	13-Aug-2018
Category:	Documents
Upload:	dokien
View:	236 times
Download:	1 times

Storage Subsystem Administration Guide - Hitachi Data … · Storage Subsystem Administration Guide...

Documents