across the silicon spectrum: xeon phi to quark – unleash the performance in your code (paul...

64
Intel Confidential Do Not Forward Across the Silicon Spectrum: Xeon Phi to Quark Unleash the Performance in Your Code Paul Butler Software Solutions Group Developer Products Division

Upload: intel-software-brasil

Post on 06-May-2015

414 views

Category:

Technology


2 download

DESCRIPTION

Paul Butler's presentation at Intel Software Day 2013 (10/22/2013) Learn how to access robust Intel resources (programs, initiatives, content, tools) available to software developers in Brazil supporting their software development life cycle across all platforms (Windows, Linux, Mac/iOS, and Android)

TRANSCRIPT

Page 1: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel Confidential — Do Not Forward

Across the Silicon Spectrum: Xeon Phi to Quark

Unleash the Performance in Your Code

Paul Butler

Software Solutions Group

Developer Products Division

Page 2: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Abstract

2

Learn how to access robust Intel resources (programs, initiatives,

content, tools) available to software developers in Brazil supporting

their software development life cycle across all platforms (Windows,

Linux, Mac/iOS, and Android)

Page 3: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel Software and Services Group

Global Footprint

~14,000+ Employees, >35 Sites

Page 4: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Now – who are you?

4

Type of developer

•Industry •Energy, academic, financial, health care/medical instrumentation, retail POS, In vehicle instruments, aerospace, gaming, defense, telecom, industrial/manufacturing

•Primary Platform •OS (Windows / Linux / Mac / Android / HTML5)

•Are you looking for cross OS support

•Architecture

X86 (Xeon Phi, Xeon, Core, Atom, Quark)

ARM

•Use of tools •Compilers, debuggers, tuning tools, libraries

•Where do you get your information / training?

•Are you familiar with Intel Software?

Page 5: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Agenda

Programs and resources • Academic Programs

• Content and Training

Technical Computing, Enterprise and HPC Development

Software • Parallelism

• Tools

• Starter Kit

Embedded System/Device Development Software • The Internet of Things (IoT)

• Tools

Game and Media Development Software • Software Development Kits (SDK)

5

Page 6: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Academic Programs

and

On Line Content

6

Page 7: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Free Software Available for Students! License Type

Student

Current students at degree-granting institutions. Free C++ tools. Discounts on Fortran tools.

Academic Researcher

For faculty or staff doing research at degree-granting

institutions. Discounted tools.

Classroom

Discounted tools to teach a curriculum at degree granting institutions.

Evaluation

Fully functional tools for any type of developer, expires in 30 days.

Contains: Intel® C++ Composer XE

• Intel® C++ Compiler • Intel® Math Kernel Library (MKL) • Intel® Threading Building Blocks (TBB) • Intel® Integrated Performance Primitives (IPP) • Intel® Cilk™ Plus

Intel® Advisor XE Intel® VTune™ Amplifier XE Intel® Inspector XE

USD $129

intel.ly/intel-education-sw 7

Page 8: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® Developer Zone Academic Community For Professors Teaching Parallelism

8

Education

Courseware

Apply for an Intel

Software Tools

Grant

software.intel.com/academic

Questions? [email protected]

Remote Manycore

Testing Lab

Page 11: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Embedded University Kit Contents

• Development Board • System CD • Quartus II CD

• Quick Start Guide • USB Cable • Power Cable • IR Remote

• Loopback Board • Intel System Studio

Intel is enabling universities with Intel Atom technology

platform kits for you to use.

11

Intel Confidential

Page 12: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Product Briefs, Evaluation Guides, White Paper

Intel Confidential

Product Briefs

Intel® Cluster Studio XE 2013

Intel® Parallel Studio XE 2013

Intel® Composer XE 2013

Intel® VTuneTM Amplifier XE 2013

Intel® Inspector XE 2013

Intel® Advisor XE 2013

Intel® Math Kernel Library 11.0

Intel® Integrated Performance Primitives 7.1 Library

Intel® Threading Building Blocks 4.1

Intel® MPI Library 4.1

Intel® Graphics Performance Analyzers 2012

Intel® SDK for OpenCL Applications 2012

Intel® Media SDK 2012

Intel® System Studio 2013 for Linux * OS

Intel® Perceptual Computing SDK

Evaluation Guides

Evaluation Guide Portal

Get an easy Performance Boost even with Unthreaded Apps

A Simple Path to Parallelism with Intel® Cilk™ Plus

Efficiently Introduce Threading using Intel® TBB 4.1

Design Parallel Performance with Less Risk and More Impact

Resolve Resource Leaks to Improve Program Stability

Eliminate Threading Errors to Improve Program Stability

Eliminate Memory Errors to Improve Program Stability

Improve C++ Code Quality with Static Analysis

Improve Fortran Code Quality with Static Analysis

White Papers

The ROI from Optimizing Software Performance with Intel Parallel Studio XE

A Concise Guide to Parallel Programming Tools for Intel Xeon Processors

Java support in Intel VTune Amplifier XE

An Introduction to Vectorization with the Intel® C++ Compiler

An Introduction to Vectorization with the Intel® Fortran Compiler

All the links on this page are public that can be accessed from the internet

Intel ® Xeon PhiTM Coprocessor

Xeon Phi MIC Developer home page

Programming for Multicore & Many-core

Xeon Phi Solution Brief: Parallel Processing, Unparalleled Discovery

Xeon Phi Webinar (Slides & Videos)

Beyond Offloading: Programming Models for Xeon Phi (IDF deck)

How to access links:

Click on link in Slideshow mode, OR

Right click on link for Hyperlink options

Page 13: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Case Studies CAE/Manufacturing

Flow Science uses Cluster Studio XE (ITAC, MPI) for CFD

Altair Speeds Complex Simulation w/ Xeon Phi

Altair uses Intel compilers and MPI for CAE and CFD software

MSC Software SimXpert w/ TBB

SIMULIA turns to Parallel Studio XE

Altair crash simulation w/ Intel SW tools

ESI Group achieves up to 450% faster performance

Financial

Black Scholes w/ Xeon Phi

Monte Carlo w/ Xeon Phi

DCSG: Thomson Reuters delivers real-time financial information

DRD: Computing Black Scholes w/ Intel AVX

Image and Video

NEC used Intel compilers for Video Conversion

Fixstars High Speed CG Renderer using Parallel Studio XE

Nik Software rendering speed of HDR by 1.3x

Envivio* video encoding w/ Parallel Studio

Medical

Massachusetts General Hospital achieves 20x increase

Education

ISPP: eLearning Software earns A+

Kyoto University: Xeon and Cluster Studio XE

Gaming and Digital Content Creation

USC Gaming students use GPA, TBB and Parallel Studio XE

Golaem uses TBB, PSXE for crowd control

Dreamworks uses Intel MKL for Dazzling Special Effects

Geomerics removes Bakeware from the Runtime using Intel GPA

All the links on this page are public that can be accessed from the internet

How to access links:

Click on link in Slideshow mode, OR

Right click on link for Hyperlink options

AEROSPACE

Aerospace Supercomputing Parallelism Advantage

Parallel Performance for University of Florence and Avio

Additional Case Studies

HPC

RWTH Aachen University adopts Parallel Studio

Comparing Arrays of Structures & Structures of Arrays on Xeon vs Xeon Phi

Page 14: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Bundled Suites

Intel® Parallel Studio XE 2013, Intel® Cluster Studio XE 2013

Getting Started Tutorial

Learn: Product Training

Threading Prototyping Tool

Intel® Advisor XE 2013

Getting Started Tutorial

Learn: Product Training

Profiler

Intel® VTune Amplifier XE 2013

Getting Started Tutorial

Learn: Product Training

Error Checking

Intel® Inspector XE 2013

Getting Started Tutorial

Learn: Product Training

Compilers

Intel® Composer XE 2013

Getting Started Tutorial for C++ and Fortran

Learn: Product Training

Libraries

Intel® Math Kernel Library 11.0

Learn: Product Training

Intel® Integrated Performance Primitives

Learn: Product Training

Programming Models

Intel® Cilk Plus

Intel® Threading Building Blocks

Intel® OpenMP*

Intel® Coarray Fortran

Intel® SDK for OpenCL Apps

Technical Computing, Enterprise & HPC Tools

All the links on this page are public that can be accessed from the internet

How to access links:

Click on link in Slideshow mode, OR

Right click on link for Hyperlink options

Page 15: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Visual Computing Tools

15

Intel Media SDK

Media SDK Homepage

Media SDK Download

Media SDK for Linux Servers homepage

Intel Perceptual Computing SDK

Product Homepage

Download

Intel SDK for OpenCL*

Intel SDK for OpenCL* Applications 2013 Homepage

Intel SDK for OpenCL* Applications XE 2013 Homepage (Xeon and Xeon Phi)

Intel Graphics Performance Analyzer

Product Homepage and Download

Getting Started Guide for Windows, Android

Beacon Mountain

Product Homepage and Download

Product Brief

Visual Computing Tools Comparison

Page 16: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Embedded and Intelligent Systems Tools

16

Intel System Studio

Home page

Product Brief

Getting Started Video

11 modules of training

All Labs and Training

Wind River

Wind River Linux Performance Studio for Intel® architecture

Wind River VxWorks*

Wind River On-Chip Debugging

Wind River Simics*

CoFluent

Intel® System Modeling and Simulation Tools from CoFluent

Additional Useful Links: Embedded Tools Home Page

Page 18: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Agenda

Programs and resources • Academic Programs

• Content and Training

Technical Computing, Enterprise and HPC Development

Software • Parallelism

• Tools

• Starter Kit

Embedded System/Device Development Software • The Internet of Things (IoT)

• Tools

Game and Media Development Software • Software Development Kits (SDK)

18

Page 19: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Technical Computing,

Enterprise and HPC

Development Software

19

Efficiently Produce

Fast, Scalable and Reliable

Applications

Page 20: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Parallel Programming?

Why do you as Developers Care?

Page 21: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Photos: Intel Crop. 21

Page 22: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Where are most these new transistors going?

Answer: Parallel HW!

Why do you care?

Answer: It’s where (dramatically) more performance is!

Page 23: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)
Page 24: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® Xeon Running Serial Code

Intel® Xeon Parallelized Code

145X FASTER

67.097 SECONDS

0.46 SECONDS

Leaving Performance on the Table?

Intel® Xeon Phi™ Parallelized Code

340X FASTER

0.197 SECONDS

Page 25: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Many verticals benefit from parallelism Intel Development Products are used here

25

Bioinformatics

Financial Analytics

Energy Medical Imaging & Analysis

Signal Processing

Engineering Design

Science & Research

Broadcast & Film

3D Modeling & Visualization

Digital Content Creation

GIS & Satellite Imagery

Database Search & Business Intelligence

Defense

& Security

Game Development

Telecommunications

Page 26: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

• Industry-leading performance from advanced compilers

• Comprehensive libraries

• Parallel programming models

• Insightful analysis tools

26

More Cores. Wider Vectors. Performance Delivered. Intel® Parallel Studio XE 2013 and Intel® Cluster Studio XE 2013

Serial Performance

Scaling Performance

Efficiently Multicore Many-core

128 Bits

256 Bits

512 Bits

61+ cores

More Cores

Wider Vectors Task & Data

Parallel Performance

Distributed Performance

Page 27: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

27

Phase Product Feature Benefit

Build

Intel®

Advisor XE Threading Prototyping Tool (Studio products only)

• Data driven threading design adds parallelism with less risk, less effort and more impact

Intel® Composer XE

• C/C++ and Fortran compilers • Intel® Threading Building Blocks • Intel® Cilk™ Plus • Intel® Integrated Performance Primitives • Intel® Math Kernel Library

• Enabling solution to achieve the application performance and scalability benefits of multicore and forward scale to many-core

Intel® MPI Library†

High Performance Message Passing (MPI) Library • Enabling High Performance Scalability, Interconnect

Independence, Runtime Fabric Selection, and Application Tuning Capability

Verify & Tune

Intel® VTune™

Amplifier XE Performance Profiler for optimizing application performance and scalability

• Remove guesswork, saves time, makes it easier to find performance and scalability bottlenecks

Intel® Inspector XE

Memory & threading dynamic analysis for code quality Static Analysis for code quality

• Increased productivity, code quality, and lowers cost, finds memory, threading , and security defects before they happen

Intel® Trace Analyzer & Collector†

MPI Performance Profiler for understanding application correctness & behavior

• Analyze performance of MPI programs and visualize parallel application behavior and communications patterns to identify hotspots

Intel® Parallel Studio XE 2013 and Intel® Cluster Studio XE 2013 †

Efficiently Produce Fast, Scalable and Reliable Applications

Page 28: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Superior C++ Compiler Performance

28

More Performance • Just recompile

• Uses Intel® AVX and Intel® AVX2 instructions

• Intel® Xeon Phi™ product family support, Linux: Compiler, debugger (Linux)

• Intel® Cilk™ Plus: Tasking and vectorization

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary.

You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. * Other brands and names are the property of their respective owners. Benchmark Source: Intel

Optimization Notice: Intel’s compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any

optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more

information regarding the specific instruction sets covered by this notice. Notice revision #20110804

Click here to view

Intel® Compiler 14.0

benchmark configuration

Page 29: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Superior Fortran Compiler Performance

29

More Performance • Just recompile • Intel® Xeon Phi™ product family: Linux compiler, debugger support • Access to Intel® AVX and Intel® AVX2 instructions (-xa or /Qxa) • Auto-parallelizer & directives to access SIMD instructions • Coarrays & synchronization constructs support parallel programming • Loop optimization directives: VECTOR, PARALLEL, SIMD • More control over array data alignment (align arrayNbytes)

Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary.

You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. * Other brands and names are the property of their respective owners. Benchmark Source: Intel

Optimization Notice: Intel’s compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any

optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more

information regarding the specific instruction sets covered by this notice. Notice revision #20110804

Click here to view

Intel® Compiler 14.0

benchmark configuration

Page 30: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Data Driven Threading Design Intel® Advisor XE – Threading Prototyping Tool for Architects

Have you:

Tried threading an app, but seen little

performance benefit?

Hit a “scalability barrier”? Performance gains

level off as you add cores?

Delayed a release that adds threading because

of synchronization errors?

Breakthrough for threading design:

• Quickly prototype multiple options

• Project scaling on larger systems

• Find synchronization errors before implementing

threading

• Separate design and implementation, design without

disrupting development

30 Add Parallelism with Less Effort, Less Risk and More Impact

Intel® Advisor XE

Page 31: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Static Analysis

Code & Security Errors

Dynamic Analysis

Memory Errors

Deliver More Reliable Applications Intel® Inspector XE and Intel® Parallel Studio XE family of suites

31

Threading Errors

Static Analysis & Pointer Checker are only available in the Parallel Studio XE family of suites. Not sold separately.

Pointer Checker

Pointer Errors

Intel®

Inspector XE

alone

Added bonus features in

Intel®

Parallel Studio XE

suites Intel compiler inspects source. Use any

compiler for production.

Intel compiler run time checks. Traps

memory accesses beyond allocated

addresses.

Intel® InspectorXE

• No special builds required. Use your normal compiler.

• Find errors that regression testing and static analysis miss.

Find errors earlier with

less effort

Page 32: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® VTune™ Amplifier XE Performance Profiler

32

Where is my application…

Spending Time? Wasting Time? Waiting Too Long?

• Focus tuning on functions taking time

• See call stacks

• See time on source

• See cache misses on your source

• See functions sorted by

# of cache misses

• See locks by wait time

• Red/Green for CPU

utilization during wait

• Windows & Linux

• Low overhead

• No special recompiles Claire Cates

Principal Developer, SAS Institute Inc.

We improved the performance of the latest

run 3 fold. We wouldn't have found the

problem without something like Intel® VTune™

Amplifier XE.

Intel® VTune™ Amplifier XE

Advanced Profiling for Scalable Multicore Performance

Page 33: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Parallelize Applications For Performance Intel® Threading Building Blocks (Intel® TBB)

A popular, proven parallel C++ template library

Scalable memory allocation

Load-balancing

Work-stealing task scheduling

Thread-safe pipeline

Flexible flow graph

Concurrent containers

High-level parallel algorithms

Numerous synchronization primitives

Open source, and portable across many OSs

33

Simplify Parallelism with a Scalable Parallel Model

Michaël Rouillé, CTO, Golaem

"Intel TBB provided us with optimized code that we did not have to develop or maintain for critical system services. I could assign my developers to code what we bring to the software table

Page 34: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® Cilk™ Plus implemented with commercial support; simplifies

going parallel

34

Enhanced performance and utilization of future Intel CPU features

SIMD pragma loops, vector length, and elemental functions support

OS X* support

cilk_for (int i=0; i<n; ++i) { Foo(a[i]);

}

Open spec at: cilkplus.org

int fib(int n) { if (n <= 2) return n; else { int x,y; x = fib(n-1); y = fib(n-2); return x+y; } }

int fib(int n) { if (n <= 2) return n; else { int x,y; x = cilk_spawn fib(n-1); y = fib(n-2); cilk_sync; return x+y; } } Turn serial code Into parallel code

Parallel loops made easy

Page 35: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Scale Performance

Intel® MPI Library 4.0 Optimized MPI application performance

Application-specific tuning

Automatic tuning

Lower latency and multi-vendor interoperability

Industry leading latency

Performance optimized support for the latest OFED capabilities through DAPL 2.0

Full conformance with the MPI 2.1 standard

Faster MPI communication

Optimized collectives

Simplify and accelerate clusters

“Intel® Cluster Ready”

More robust MPI applications

Seamless interoperability with Intel® Trace Analyzer and Collectorbvb

35

Page 36: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Scale Performance - Tune Hybrid Cluster MPI and Thread

Performance

36

Tune cross-node MPI

Visualize MPI behavior

Evaluate MPI load balancing

Find communication hotspots

Intel®

Trace Analyzer and Collector

Page 37: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Demand: strategic marketing buzz begins to pay dividends

78% of HPC users plan to buy coprocessor with next server* (Source: IDC)

Phi leads all accelerators in user pref for future purchase (Source: IDC)

Consistent interest from academics and developers (Source: Intel Field)

Problem: cost and complexity remain barriers to adoption

Solution: Entry Point Priced Xeon Phi starter kit program through partners

Xeon Phi Starter Kit Overview

+ + + +

Top 10 List

webinar,

simple labs,

forum, &

more

Starting below $5K

Intel Kit System

+

* Over 2x the percentage from 2011 survey

Page 38: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Agenda

Programs and resources • Academic Programs

• Content and Training

Technical Computing, Enterprise and HPC Development

Software • Parallelism

• Tools

• Starter Kit

Embedded System/Device Development Software • The Internet of Things (IoT)

• Tools

Game and Media Development Software • Software Development Kits (SDK)

38

Page 39: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Embedded

System/Device

Development Software

39

Create fast, efficient embedded

and mobile systems in less time.

Page 40: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

The Internet of Things

50,000,000,000 Number of devices connected to the Internet by 2020

Connected embedded “ Intelligent Systems“

Not only PCs, smartphones and tablets but cars,

houses, machines, sensors, trees, parking meters, thermostats, doors, windows, cardiac monitors, tires, roads, cars, supermarket shelves, watches, glasses, and much more.

Page 41: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Connected - Smart World

Page 42: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel Confidential — Do Not Forward

EXTREME Inclusion

42

Key Industries To Watch For

Medical Instrumentation

In-vehicle-instruments

Industrial / manufacturing

Communications Military

Aerospace applications

Retail POS

Embedded Customers: Consumer/Non-consumer Electronics, Medical Instrumentation,

Aerospace/Military, Automotive/Transportation, Manufacturing / Process Control

Challenges for Developers

As new compute platforms include more IP blocks, more cores and other coprocessors,

embedded systems developers need to ensure that their software stack is power

optimized and best performs and scales on new hardware while meeting TTM needs

Page 43: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

System Software Developer Challenges

Meeting release schedule

System reliability

Power efficiency &

application performance

8%

8%

15%

16%

22%

Microprocessor

Programming Tools

Schedule

Engineering Team/Skill levels

Debugging Tools

If you could improve one thing about your embedded design activities, what

would it be?

UBM Electronics - 2012

Embedded Market Survey VDC Research – Strategic Insights 2012:

Embedded Software & Tools Market.

October 2012

Page 44: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® System Studio Deep System Insights for Embedded and Mobile Developers

44

Accelerate Time To Market

Speed-up development and testing with deep

hardware and software insights

Strengthen System

Reliability

Enhance code stability

using in-depth system-wide

debuggers and analyzers

Boost Power Efficiency and Performance

Boost system power efficiency and performance using

system-wide analyzers, compilers

and libraries

Page 45: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® System Studio – What is it?

“Wind River is excited about the release of Intel® System Studio which includes support for the Wind River Linux operating system”,

said Dinyar Dastoor, VP Product Management, Operating Environments at Wind River. “Providing deep insights into Intel® Architecture, these tools help embedded and system developers rapidly implement Wind River Linux solutions that are reliable, responsive and power efficient.”

Debuggers

System Application

Analyzers

Power & Performance Memory & Thread errors

Compiler and Libraries

C/C++ Compiler Signal, Media, Data & Math Processing

System & Application Code running Linux* JTAG

Interface

Embedded or Mobile System

Integrated software tool suite that provides deep system-wide insights to help:

Accelerate Time To Market

Strengthen System Reliability

Boost Power Efficiency and Performance

Page 46: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Agenda

Programs and resources • Academic Programs

• Content and Training

Technical Computing, Enterprise and HPC Development

Software • Parallelism

• Tools

• Starter Kit

Embedded System/Device Development Software • The Internet of Things (IoT)

• Tools

Game and Media Development Software • Software Development Kits (SDK)

46

Page 47: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Game and Media

Development Software

47

Immersive interactivity for

multimedia apps and more

Page 48: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

48

Page 49: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

49

Page 50: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Comprehensive Android Development Environment For ARM* & Intel® Atom™ Smartphones, Tablets Tools for design, code, debug & app acceleration Helps Shorten the Android App Development Cycle

Intel Tools Intel® Hardware Accelerated Execution Manager Intel® Graphics Performance Analyzers System

Analyzer Intel® Integrated Performance Primitives Preview Intel® Threading Building Blocks Intel® Software Manager

Third-Party Tools Google* Android SDK (ADT Bundle) Android NDK Eclipse* Integrated Development Environment Android Design Cygwin* (for Microsoft* Windows operating

systems)

Free download at:

intel.com/software/BeaconMountain

A Development Environment for Native Android* Apps on ARM* & Intel® Atom™ Devices

Beacon Mountain Preview v0.5

50

Page 51: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

51

Page 52: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel Confidential — Do Not Forward

Hardware Acceleration for Video

Intel® Media SDK 2013 for Servers

*Other names and brands are the property of their respective owners

Deliver Streaming Density, Power Efficient Transcode

Future Proof: Develop now for today and tomorrow's server platforms

Save Time and Money: Spend more time delivering content

Page 53: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Information: intel.com/software/mediasdk/servers

Intel® Media SDK 2013 for Servers Hardware Accelerator for Video

Develop hardware accelerated video workloads on Intel® Xeon E3-1285Lv3,

3rd and 4th generation Intel Core™ platforms with Intel HD Graphics

OS: Ubuntu* 12.04 LTS, SUSE Linux Enterprise Server* (SLES) 11 SP2, Windows

Server 2012

CODECS: H.264, MPEG-2, VC-1

Encode, decode, and transcode for server-based real-

time streaming, teleconferencing, and video analytics

Delivers more concurrent HD transcodes, streaming

density at significantly lower TCO

*Other names and brands are the property of their respective owners

Page 54: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Interaction With Computing Devices

What’s Next?

Face

Tracking Local

Speech Recognition

Augmented Reality

Finger/Gesture Tracking

Background Subtraction

Page 55: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® Perceptual Computing SDK Bring Your Apps into the Next Dimension

An advanced and comprehensive SDK for NUI

Download the SDK or Order the Camera at: intel.com/software/perceptual

2013 SDK: FREE Download

Perceptual Usage Modes Supported:

Close-range finger and hand depth tracking Face Analysis, Tracking Speech Recognition Augmented Reality

Creative* Interactive Gesture Camera Developer Kit available for purchase

@ $149 HD Webcam IR Depth Sensor Dual-array microphones

1Intel Perceptual Computing is supported on 2nd, 3rd and 4th generation Intel Core

Processors

Page 56: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® Graphics Performance Analyzer A Suite of graphics analysis and optimization tools to help game developers make games and other

graphics-intensive applications run even faster.

Optimization Notice: : Intel’s compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Heads-Up

Display

Stand-

Alone

System Analyzer Heads Up Display (HUD) & Stand-Alone Modes

Real-time, in-game analysis with graphical metrics

displays and state overrides

Frame Analyzer Deep frame performance analysis down to draw call level, including shaders, textures, D3D states, pixel history, and textures

Platform Analyzer

Full system analysis of CPU metrics and workloads across multiple threads and cores, plus simultaneous GPU metrics

System Analyzer (Stand-alone) for Android

Performance, Power Profiling: CPU, graphics API level, GPU hardware & power metrics

Page 57: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® SDK for OpenCL* Applications

Intel Confidential 57

Develop Cutting-Edge Visual Computing Applications Get Maximum Processing Power from the Platform For 3rd and Future 4th Gen Intel® Core™ Processors For Both CPUs and Intel HD Graphics

Develop Cutting–edge Visual Computing Applications for media & image processing, video

editing, playback & transcoding , digital & 3D content creation, visualization

Develop with an Open Standard Compute Model Fully conformant with OpenCL 1.2 to enable applications with cross architecture functional portability

FREE Download: intel.com/software/opencl

Run Applications on Windows 7 or 8 OS

Ensuring compatibility on a wide range of desktop applications

Develop With Tools You Know , Tools You Need Including Microsoft

Visual Studio IDE, Microsoft DirectX *, OpenGL*, Intel Media SDK, Intel vTune Amplifier XE, Intel GPA,

Kernel Builder, Kernel Debugger

Page 58: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

The Market

Page 59: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel is helping software developers around the globe to create and deploy incredible cross-platform

experiences, reach more customers, reduce costs and improve time-to-market

• Intel® HTML5 Developer Zone provides everything

to learn, develop, test and deploy true cross-

platform HTML5 apps

• Intel® XDK, a no cost, integrated and front-to-back

HTML5 App Dev Environment for True Cross-

Platform Apps for multiple App Stores, and form

factor devices

• Intel® HTML5 App Porter Tool – Beta, an iOS* app

to HTML5 source-code and API converter to

broaden the market reach for iOS developers

• Intel contributes to Open Source, Standards, and

provides Tools & Resources to support HTML5

Let HTML5 Thrive…

59

Page 60: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

App Dev

Center Intel®XDK

Integrated and front-to-back HTML5 App Dev Environment for True Cross-Platform Apps for multiple

App Stores, and form factor devices

• Intel cloud-based build system: packaged hybrid apps

iOS*, Android*, Windows* 8, Windows Phone* 8, Tizen*

• Packaged web apps for Chrome* Stores, Facebook*, and Web

apps for self-hosting

• No need to download Native Platform SDKs

• Developer Frontend • Editor, Device Emulator, Debugger, • App Preview - On-device Testing • App Framework (JQMobile and UI) • App Game Interfaces • Integrated Intel® App Porter Tool

(Windows* only) • “New App Experience” quick-start wizard • Hosted on Windows*, MacOS X*, runs in

Chrome*

Intel® XDK

Page 61: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel® XDK Intel® XDK enables software developers to develop, test and deploy HTML5 web and hybrid apps

across platforms, app stores and multiple form factors

61

Write Once, Deploy Everywhere HTML5 Web Apps, Hybrid Apps

Amazing App Experience Optimized UI/UX JS Libraries

Native Themes Responsiveness

PRO UI Builder

Brackets* HTML5 Editor Prototyping App Builder iOS* to HTML5 App Porter Tool Cordova* Device Emulator On-The-Fly On-Device App Tester

Cloud-based hybrid app build system

Faster-Time-To-Market Integrated Front-To-End Tools Solution

Short Learning Curve Simplified Workflow

Step-by-step process

Samples PRO UI Builder

Native App Stores WebApp Stores

The Intel® XDK is free at http://xdk.intel.com

Page 62: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

NEW: Intel C++ for Android

Available for Limited FREE

Download:

Intel Confidential 63

Page 63: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

Intel Confidential — Do Not Forward

Page 64: Across the Silicon Spectrum: Xeon Phi to Quark – Unleash the Performance in Your Code (Paul Butler)

INFORMATION IN THIS DOCUMENT IS PROVIDED “AS IS”. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE,

TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. INTEL ASSUMES NO LIABILITY WHATSOEVER

AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO THIS INFORMATION INCLUDING LIABILITY OR

WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY

PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.

Software and workloads used in performance tests may have been optimized for performance only on Intel

microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems,

components, software, operations and functions. Any change to any of those factors may cause the results to vary. You

should consult other information and performance tests to assist you in fully evaluating your contemplated purchases,

including the performance of that product when combined with other products.

Copyright © , Intel Corporation. All rights reserved. Intel, the Intel logo, Xeon, Xeon Phi, Core, VTune, and Cilk are

trademarks of Intel Corporation in the U.S. and other countries. Optimization Notice

Intel’s compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice.

Notice revision #20110804

Legal Disclaimer & Optimization Notice

Copyright© 2012, Intel Corporation. All rights reserved.

*Other brands and names are the property of their respective owners.

65