Copyright © 2008 Intel Corporation. All rights reserved. Intel Delivering Leadership HPC Technology – today and tomorrow – …for Grids …for Grids Sept 22th,

Copyright © 2008 Intel Corporation. All rights reserved.

Intel Delivering Leadership HPC Intel Delivering Leadership HPC Technology Technology – – today and tomorrow – today and tomorrow –

……for Gridsfor Grids Sept 22th, 2008Sept 22th, 2008EGEE 08 ConferenceEGEE 08 ConferenceIstanbulIstanbul

Stephan GillichStephan GillichDirector HPCDirector HPCEMEA Enterprise MarketingEMEA Enterprise MarketingIntel GmbHIntel GmbH


Legal Disclaimer INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL® PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY

ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL® PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. INTEL PRODUCTS ARE NOT INTENDED FOR USE IN MEDICAL, LIFE SAVING, OR LIFE SUSTAINING APPLICATIONS.

Intel may make changes to specifications and product descriptions at any time, without notice. All products, dates, and figures specified are preliminary based on current expectations, and are subject to change without notice. Intel, processors, chipsets, and desktop boards may contain design defects or errors known as errata, which may cause the product to

deviate from published specifications. Current characterized errata are available on request. This document may contain information on products in the design phase of development. The information here is subject to change

without notice. Do not finalize a design with this information. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined." Intel reserves

these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them.

Intel Corporation may have patents or pending patent applications, trademarks, copyrights, or other intellectual property rights that relate to the presented subject matter. The furnishing of documents and other materials and information does not provide any license, express or implied, by estoppel or otherwise, to any such patents, trademarks, copyrights, or other intellectual property rights.

Wireless connectivity and some features may require you to purchase additional software, services or external hardware. Nehalem, Penryn, Westmere, Sandy Bridge and other code names featured are used internally within Intel to identify products that are in

development and not yet publicly announced for release. Customers, licensees and other third parties are not authorized by Intel to use code names in advertising, promotion or marketing of any product or services and any such use of Intel's internal code names is at the sole risk of the user

Performance tests and ratings are measured using specific computer systems and/or components and reflect the approximate performance of Intel products as measured by those tests. Any difference in system hardware or software design or configuration may affect actual performance.

Intel, Intel Inside, Pentium, Xeon, Core and the Intel logo are trademarks of Intel Corporation in the United States and other countries. *Other names and brands may be claimed as the property of others. Copyright © 2008 Intel Corporation.


Agenda

Intel HPC and Grid activities Leading technology today …. And tomorrow Enabling Grid and beyond


Intel in the Grid

Intel is driving HPC technology and Network technology

Intel Volume - HPC technology enables Clusters Clusters and Networks enable Grids Intel has direct involvement in the Grid area


Intel in High Performance Computing

A Long Term Commitment to HPC SolutionsA Long Term Commitment to HPC Solutions

Dedicated,renowned expertise

Large scaleclustersfor test and optimization

Broad SW tools portfolio

Teraflopsresearch

Leading performance, performance/watt

DefinedHPCapplicationplatform

Platform buildingblocks


Intel’s Grid Activities

Develop prototype Grid components– Simplify creation of portable Grid applications

– Grid Programming Environment (GPE)

– GridBeans encapsulate application user interfaces

– Enables management and use of virtualized Grid resources (CERN cooperation)

– Further development by the Unicore family of projects (FZ Juelich)

Engage with Grid experts and users– European–union funded R&D projects (UniGrids,

NextGRID, SimDAT)

Interact with ecosystem, standards bodies– Open Grid Forum (technical influence, board membership)


IntelArchitecture

Core

New Materials and DesignsNew Materials and Designs

Single Core EnhancementsSingle Core Enhancements

Platform EnhancementsPlatform Enhancements

Scaling technology forwardScaling technology forward

Multi to Many-Core Multi to Many-Core


IntelArchitecture

Core

Tri-Gate, Nanotubes Tri-Gate, Nanotubes

MMX SSE AVX MMX SSE AVX

PCIe, IMC, QPI, SOC PCIe, IMC, QPI, SOC

Scaling technology forwardScaling technology forward

Dual Quad …Dual Quad …


2007

PenrynPenryn NehaleNehalemm

Sandy Sandy BridgeBridge

WestmereWestmere

NEWMicroarchitecture

45nm


32nm

Compaction/Derivative

45nm45nm

Compaction/Derivative

32nm32nm

Tick/Tock: Our Model for Sustained Tick/Tock: Our Model for Sustained Microprocessor Leadership Microprocessor Leadership

2006

IntelIntel®® Core™Core™


65nm

Forecast

Future options subject to change without notice.

2008


Scaling Performance Forward

Achieving Moore’s Law Achieving Moore’s Law Through The Foreseeable Through The Foreseeable FutureFuture

Write traditional software Write traditional software (FTN, C, C++) (FTN, C, C++) Intel software Intel software tools : tools : Compile Compile onceonce, generating , generating optimized, optimized, native native IAIA code code

Balanced Computing: Balanced Computing: – Processor Capacity, Memory Processor Capacity, Memory

Capacity, Bandwidth CapacityCapacity, Bandwidth Capacity

11

Scale Performance Forward To Multi & Many Core Solutions

11

………………

………………


Larrabee is Intel’s first many core Larrabee is Intel’s first many core processorprocessor

First Larrabee product targets First Larrabee product targets enthusiast graphicsenthusiast graphics

Each core is a complete Intel processorEach core is a complete Intel processor

Efficient inter-block communicationEfficient inter-block communication

Throughput Computing: Throughput Computing: A Many Processor Architecture. LARRABEEA Many Processor Architecture. LARRABEE

VECTORIA COREVECTORIA CORE

INTERPROCESSOR NETWORKINTERPROCESSOR NETWORK

INTERPROCESSOR NETWORKINTERPROCESSOR NETWORK

FIX

ED

FU

NC

TIO

N L

OG

ICFI

XE

D F

UN

CTIO

N L

OG

IC

MEM

OR

Y a

nd I/O

IN

TER

FAC

ES

MEM

OR

Y a

nd I/O

IN

TER

FAC

ES








COHERENTCACHE

COHERENTCACHE

……

……

……

……

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

COHERENTCACHE

1212


The Challenge The Challenge Parallel ProgrammingParallel Programming

?Irregular Patterns, Data Structures and Serial Algorithms

Increasing Cores Vector Instructions Cache and Interconnect Latency

Scale to Multi-Core Today → HardScale to Many-Core Tomorrow → Harder

13


Today: IntelToday: Intel®® Software Tools for Parallelism Software Tools for Parallelism

Introduce ParallelismIntroduce Parallelism

Confidence / CorrectnessConfidence / Correctness

Optimize / TuneOptimize / Tune

Architectural AnalysisArchitectural Analysis

*Other names and brands may be claimed as the property of others.


Research: Ct: A Throughput Programming Research: Ct: A Throughput Programming LanguageLanguage

1 1 0 00 1 0 1 0 1 0 00 0 1 1

1 1 0 00 1 0 1 0 1 0 00 0 1 1+

Thread 4

0 0 1 1

0 0 1 1+

Thread 3

0 1 0 0

0 1 0 0+

Thread 2

0 0 0 1

0 0 0 1+

Thread 1

1 1 0 1

1 1 0 1+

Ct JIT Compiler:

Auto-vectorization, SSE, AVX, LarrabeeCore 1

SIMD Unit

Core 2

SIMD Unit

Core 3

SIMD Unit

Core 4

SIMD Unit

Programmer Thinks Serially; Ct Exploits Parallelism

Ct Parallel Runtime:

Auto-Scale to Increasing Cores

User Writes Serial-LikeCore Independent C++ Code

TVEC<F32> a(src1), b(src2);

TVEC<F32> c = a + b;

c.copyOt(dest);

TVEC<F32> a(src1), b(src2);

TVEC<F32> c = a + b;

c.copyOt(dest);

Primary Data Abstraction is the Nested Vector

Supports Dense, Sparse, and Irregular Data


A practical case: CERN – Chasing The Intractable

Ventures into the subatomic world – latest experiment: LHC

Data Analysis Grid for LHC – Revolutionizes the way scientists

share and analyze data

Openlab collaboration essential for building the IT infrastructure

Intel contributes in a unique technology collaboration since 2003

Science At The Edge Meets Technology At The Edge


Beyond: Cloud computing

Intel, HP, Yahoo create cloud-computing labs29 Jul 2008 17:22A global cloud infrastructure for researchers will help advance the technology, according to the companiesHP, Intel and Yahoo, in conjunction with academic and other institutions around the world, have announced the creation of a global, experimental environment for cloud computing. The Cloud Computing Test Bed will initially comprise of six 'Centers of Excellence', with the University of Illinois at Urbana-Champaign (UIUC), the Karlsruhe Institute of Technology in Germany and the Infocomm Development Authority of Singapore joining in with the three companies.

Intel, HP, Yahoo create cloud-computing labs29 Jul 2008 17:22A global cloud infrastructure for researchers will help advance the technology, according to the companiesHP, Intel and Yahoo, in conjunction with academic and other institutions around the world, have announced the creation of a global, experimental environment for cloud computing. The Cloud Computing Test Bed will initially comprise of six 'Centers of Excellence', with the University of Illinois at Urbana-Champaign (UIUC), the Karlsruhe Institute of Technology in Germany and the Infocomm Development Authority of Singapore joining in with the three companies.

The next step– Grid “grown up” in

the academic / research community (also because of EGEE)

– In the commercial areas “Cloud Computing” is one of the big areas of development

– Intel is again engaged in this Area


Summary

Cluster of COTS HW important for GridsCluster of COTS HW important for Grids•Moore’s Law Continues To Double Transistors Moore’s Law Continues To Double Transistors

Every 18 MonthsEvery 18 Months•More Core’s, More Cache, More Memory, More More Core’s, More Cache, More Memory, More

Bandwidth Delivers PerformanceBandwidth Delivers Performance•Intel Software Tools Enable You To scale Intel Software Tools Enable You To scale

Performance ForwardPerformance Forward•Intel’s leading technology enables Grids and we Intel’s leading technology enables Grids and we

are actively contributingare actively contributing


HPC @ Intel

Putting it all togetherPutting it all together

Have a successful conferenceHave a successful conference

Documents

Copyright © 2008 Intel Corporation. All rights reserved. Intel Delivering Leadership HPC Technology – today and tomorrow – …for Grids …for Grids Sept 22th,