Date post: | 17-Jun-2015 |
Category: |
Technology |
Upload: | inside-analysis |
View: | 76 times |
Download: | 1 times |
Grab some coffee and enjoy the pre-show banter before the top of the hour!
The Briefing Room
Fact or Fiction: Why Do We Need In-Memory Computing?
Twitter Tag: #briefr
The Briefing Room
! Reveal the essential characteristics of enterprise software, good and bad
! Provide a forum for detailed analysis of today’s innovative technologies
! Give vendors a chance to explain their product to savvy analysts
! Allow audience members to pose serious questions... and get answers!
Mission
Twitter Tag: #briefr
The Briefing Room
Topics
This Month: BIG DATA
March: CLOUD
April: BIG DATA
2014 Editorial Calendar at www.insideanalysis.com/webcasts/the-briefing-room
Twitter Tag: #briefr
The Briefing Room
Big Data
Twitter Tag: #briefr
The Briefing Room
Analyst: Robin Bloor
Robin Bloor is Chief Analyst at The Bloor Group
[email protected] @robinbloor
Twitter Tag: #briefr
The Briefing Room
GridGain
! GridGain offers a Java-based software stack for real time Big Data processing
! Its architecture is based on a high performance in-memory processing platform that integrates compute and in-memory data grids
! The GridGain stack includes In-Memory Database, HPC and Streaming, as well as Accelerators for Hadoop and MongoDB
Twitter Tag: #briefr
The Briefing Room
Guest: Nikita Ivanov
Nikita Ivanov is founder and CTO of GridGain Systems. Nikita has led GridGain to develop advanced and distributed in-memory data and computational grid technologies. He has over 20 years of experience in software application development, building HPC and middleware platforms, contributing to the efforts of other startups and notable companies including Adaptec, Visa and BEA Systems. Nikita was one of the pioneers in using Java technology for server side middleware development while working for one of Europe’s largest system integrators in 1996. He is an active member of Java middleware community, contributor to the Java specification.
In-Memory Computing:!Facts & Myths
www.gridgain.com #gridgain
What is In-Memory Computing?
In-Memory Computing uses high-performance, distributed memory systems to compute and transact on large-scale data sets in real-time - orders of
magnitude faster than disk-based systems.
Why Now?
Paradigm Shift à la 1970s
1970s: Era of Disk!> IBM released “Winchester” IBM 340 disk
Tapes start to decline!> SQL
Era of Structured Data
2010s: Era of Memory!> 64-bit CPUs + DRAM prices drop 30% YoY
HDDs start to decline!> NoSQL + SQL
Era of Unstructured Data!
> Last frontier for storage!
RAM is a new disk, disk is a new tape.
Memory First vs. Disk First
> Disk First Architecture: 1970-2000sDisk as primary storage, memory for caching Reading Record: API call <-> OS I/O <-> I/O controller <-> disk Latency: milliseconds!
!
> Memory First Architecture: since 2000sMemory is primary storage, disk for backupsReading Record: API call <-> pointer arithmeticLatency: nanoseconds to microseconds
Myth #1: Too Expensive
Facts:!> 2013: 1TB DRAM cluster ???!
> 2015: 1TB DRAM cluster ???30% reduction YoY!
> Memory Channel Storage (MCS) NAND in DRAM form factor, 2x speed of flash, same price as flash!
> Storage Class Memory (SCM)~10x slower than DRAM, Flash price, non-volatile
Myth #1: Too Expensive
Facts:!> 2013: 1TB DRAM cluster ~$25K!
> 2015: 1TB DRAM cluster ~$10K 30% reduction YoY!
> Memory Channel Storage (MCS) NAND in DRAM form factor, 2x speed of flash, same price as flash!
> Storage Class Memory (SCM)~10x slower than DRAM, Flash price, non-volatile
Myth #2: Not Durable
Facts:!> IMC have durable backups and disk storage
Active or passive replicas, transactional read-through and write-through!> Mature IMC provide tiered storage
DRAM - Local Swap - RDBMS/HDFS!
> Operational vs. Historical datasets99% of operational datasets < 10TB
Myth #3: Flash Is Fast Enough
Facts:!> Flash on PCI-E is still... a block device.
Still going through OS I/O, I/O controller, marshaling, buffering.
Myth #4: Only For Caching
Facts:!> Caching is important use case for yesterday
Easiest adoption and a “low-hanging fruit”!> In-Memory Data Grids & HPC for today
Main system of records moving to in-memory!> Vertical and PnP products are the future
Minimal integration, maximum benefit
Twitter Tag: #briefr
The Briefing Room
Perceptions & Questions
Analyst: Robin Bloor
Hierarchical Memory
u On chip speed v RAM • L1(32K) = 100x • L2(246K) = 30x • L3(8-20Mb) = 8.6x
u RAM v SSD • RAM = 300x
u SSD v Disk • SSD = 10x
It’s Over for Spinning Disk
u SSD is now on the Moore’s Law curve
u Disk is not and never was on that curve (seek time)
u Traditional DBMS were built for spinning disk
u We may now have to rethink the DBMS…
In-Memory Is Different…
The Operational Intelligence Dynamic
We used to think of BI as “acceptable even when slow”
That was before the trend toward Operational Intelligence
In-Memory Will Disrupt
u In-memory processing is currently an accelerator but it will become the norm
u It will be used according to application value
u Latency improvements up to THREE ORDERS OF MAGNITUDE are possible
u Operational Intelligence is a natural candidate
u What are the primary applications where GridGain is being employed?
u What is the biggest server grid currently deployed by a GridGain customer? How much memory?
u How does “in-memory fault tolerance” work?
u What changes occur when memory becomes the prime data store?
u How difficult is it to develop applications in the GridGain environment in practice?
u Are any companies adopting this technology strategically?
u Are there any products that target the GridGain environment (in-memory databases, in-memory ESBs, etc.)?
Twitter Tag: #briefr
The Briefing Room
Twitter Tag: #briefr
The Briefing Room
Upcoming Topics
www.insideanalysis.com
2014 Editorial Calendar at www.insideanalysis.com/webcasts/the-briefing-room
This Month: BIG DATA
March: CLOUD
April: BIG DATA
Twitter Tag: #briefr
The Briefing Room
THANK YOU for your
ATTENTION!