Keynote (Nandini Ramani) - The Role of Java in Heterogeneous Computing & How You Can Help - by ...

Copyright © 2013, Oracle and/or its affiliates. All rights reserved. 1

The role of Java™ in Heterogeneous Computing …and how you can help! Nandini Ramani Vice President, Java Platform @eyeseewaters


©2013 Oracle Corporation

The following is intended to outline our general product direction. It is intended for information purposes only, and may not be incorporated into any contract. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions.

The development, release, and timing of any features or functionality described for Oracle’s products remains at the sole discretion of Oracle.


Java


5 of Top 5 OEMs ship Java ME #1 Development platform

9 Million Java developers worldwide

3 Billion mobile phones run Java

#1 Choice for developers

80% of mobile developers use Java 100% of Blu-Ray players ship with Java 97% of enterprise desktops run Java 89% of desktops in USA run Java 7 Billion Java Cards sold 5 Billion Java Cards in use 115 Million TV devices run Java


Hello.java javac Hello.jar Hello.class Java Virtual

Machine

Life of Java Program


Java ME Embedded

Oracle Java Platforms Overview

Java Card

SECURITY SMALL EMBEDDED MEDIUM EMBEDDED LARGE EMBEDDED DESKTOP SERVER

50KB-1MB

1MB-10MB

10MB-100MB

Footprint

Java SE Embedded

100MB+

Java SE


Rich Tools Ecosystem


Rich Tools Ecosystem

Java has rich and competitive tools ecosystem Ports to many platforms and tool chains

Choice is good


Java is Open!


OpenJDK

“The place to collaborate on an open-source implementation of the Java Platform, Standard Edition, and related projects.”

In a Nutshell

http://openjdk.java.net

http://openjdk.java.net


30+ Projects

Java 7 RI Binaries

Open Community – GPL v2 + Classpath Exception

OpenJDK Virtuous Cycle

Java for IBM Platforms

Java for SAP Platforms

Customized VM for Internal Use

...and many more

Both Gratis and Commercial offerings (Including Support, Enterprise tooling, etc, available)

Java for Red Hat Platforms

Source contributions

Companies AND Individuals


Porting and optimizing Oracle JDK for Linux ARM/V8 GA - H1 2015

64-bit


Java and Heterogeneous Compute


“The HSA Foundation goal is to make it dramatically easier to program heterogeneous parallel devices.”




Parallel

Power usage is leading to parallel throughput Programmers need to execute a mix of parallel and sequential tasks Java remains the tool of choice as hardware tilts toward the parallel

Java and Heterogeneous Computing


2002 … 2004 … 2006 …. 2011 … 2014

1.4 5.0 6 7 8 java.lang.Thread

java.util.concurrent (jsr166)

Fork/Join Framework (jsr166y)

Project Lambda History of Parallelism Support in Java

Phasers, etc (jsr166)


Parallelism Support in Java 8 – Lambdas Simplifying the code to write parallel actions

// The OLD way. Simplistic, Brittle, and Serialized // Find Total Salary of all California Employees public static int californiaSalary(List<Employee> employees, int salary) { int sum=0; for (Employee e : employees) { if (e.getStateCode() == CALIFORNIA) sum += e.getSalary(); } return sum; }


Parallelism Support in Java 8 – Lambdas Simplifying the code to write parallel actions

// Total Salary of all California Employees int sum = employees.parallelStream() .filter(e -> e.getStateCode() == CALIFORNIA) .mapToInt(e -> e.getSalary()) .sum();


Parallelism Support in Java 8 – Arrays Simplifying the code to write parallel actions

// New Parallel Methods added to Array // In place cumulation Array.parallelPrefix(anArray, BinaryOperator) // Sort Array.parallelSort(anArray); // In place mutations Array.parallelSetAll(anArray, UnaryOperator) // Special Iterator for decomposition in addition to single-element Array.spliterator(anArray)




Heterogeneous

Future is power-constrained Tradeoff is latency vs. throughput Typical system will be a few low latency cores, many high throughput cores, some special accelerators



New Challenges – Array-wise programming concepts

Java is a concurrent language with concurrent and parallel programming support built into the core

Exploit the hardware Array-wise programming

Java JDK 8 includes parallel decomposition and programming tools out of the box.



New Challenges – “WORA” Becomes Local

Transitioning from write once, run anywhere “WORA” to a world where one JVM can access multiple ISAs at once


JVM 1 OS 1 ISA 1 Chip 1 Mem 1

JVM 2 OS 2 ISA 2 Chip 2 Mem 2

JVM OS

ISA 1 Chip 1 Mem 1

ISA 2 Chip 2 Mem 2


Some cores optimized for: – High Throughput – Low latency – Low Power – Special Accelerators

Where should work be routed?

JVM OS

ISA 1 Chip 1 Mem 1

ISA 2 Chip 2 Mem 2

New Challenges – Route Work to Best Resource Java and Heterogeneous Computing

?


New Challenges – Non-Uniform Memory Access

Heterogeneous systems have bumpy costs with non-uniform memory

– JVM can manage memory on the fly in response to program dynamics

– Manage memory access and data flow across ISAs


JVM

OS

ISA 1

Chip 1

Mem 1

ISA 2

Chip 2

Mem 2 Memory


How do we meet these and other challenges?

JVM

OS

ISA 1

Chip 1

Mem 1

ISA 2

Chip 2

Mem 2


How do we meet these and other challenges?

http://openjdk.java.net/projects/sumatra/


Project Sumatra

OpenJDK Project with leadership from AMD and Oracle Goal is to simplify Heterogeneous Compute (via multiple ISAs) for

nine million+ Java developers Oracle team includes Labs and SPARC expertise HSA Standards are important:

– HSAIL helps with common back end notation – HSA helps with portability




Gary Frost Software Fellow


Histogram of male/female (5150) names Found in selected works of ‘Charles Dickens’ So conceptually we have sequence of nested loops.

Sumatra Demo

for each selected book{ for each name{ for each occurrence of ‘name’ in ‘book’{ name.count++; } } }


Can be done in parallel !


Here is the actual Java 8 code at the heart of the demo

Sumatra Demo

Arrays.stream(library) .filter(Book::isSelected) .forEach(book-> Arrays.stream(names) .parallel() // <-- Loop over the names in parallel. .forEach(name -> // for each occurrence of ‘name’ in ‘book’ // name.count++; ); );



Java 8 can efficiently dispatch the enclosed lambda across multiple cores when the stream is marked as parallel()

A Sumatra JVM will dispatch to CPU and HSA enabled GPU cores.

– JVM converts (JITS) bytecode to HSAIL and executes via HSA Runtime

Sumatra Demo

Arrays.stream(T[] array) .parallel() .forEach(t -> /* lambda */)




HSA enablement of JAVA

CPU ISA GPU ISA

JVM

Java Application

GPU CPU

OpenCL™

OpenCL™ Compiler and Runtime

APARAPI API

Java 7 OpenCL enabled Aparapi • AMD initiated Open Source project • Java API for data parallel algorithms • No need to learn OpenCL • No explicit buffer transfers

CPU ISA GPU ISA

JVM

Java Application

GPU CPU

HSAIIL

HSA Finalizer & Runtime

APARAPI / Lambda API

Java 8 HSA enabled Aparapi

• Java 8 Lambda based API

• HSA and SVM enablement

Beyond Java 8 Sumatra • Adds Native GPU acceleration to JVM • Java 8 Lambda/Stream API • JVM generates HSAIL via JIT • JVM dispatches to CPU or GPU

depending on workload characteristics.

GPU ISA

JVM

Java Application

GPU CPU

HSAIL

HSA Finalizer & Runtime

Java JDK Stream + Lambda API

JIT

CPU ISA


How you can help!

Project Sumatra is Open Source We can always use help:

– Test builds – Track progress – Are you a VM or Compiler super star? We

can always use more help!


http://openjdk.java.net/projects/sumatra/



Date post:	10-May-2015
Category:	Technology
Upload:	amd-developer-central
View:	3,170 times
Download:	0 times

Keynote (Nandini Ramani) - The Role of Java in Heterogeneous Computing & How You Can Help - by ...

Technology