Bending The Curve

Today’s presentations contain forward-looking statements. All statements made that are not historical facts are subject to a number of risks and uncertainties, and actual results may differ materially. Please refer to our most recent Earnings Release and our most recent Form 10-Q or 10-K filing for more information on the risk factors that could cause actual results to differ. If we use any non-GAAP financial measures during the presentations, you will find on our website, intc.com, the required reconciliation to the most directly comparable GAAP financial measure. 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. 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. Buyers should consult other sources of information to evaluate the performance of systems or components they are considering purchasing. For more information on performance tests and on the performance of Intel products, reference www.intel.com/software/products. 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.

Legal Information

http://www.intel.com/software/products

Bending The Curve Daryan Dehghanpisheh

May 24, 2012

Source: All trademarks and brands are registered property of their respective companies.

A Shameless Plug

3

Join in the discussion!

All trademarks and brands are registered property of their respective companies.

The “So What?”

• We have “Knowns, unknowns, and known unknowns”

• Radical performance is here. And, affordable for most.

• Dealing with data requires a whole new strategy.

• The Cost Curves of the business are bending.

All told: A new breed of competitor lurks.

4

KNOWNS, UNKNOWNS, KNOWN UNKNOWNS

What I hear about, all the time.

6 Source: 1. Google Images search, public domain All trademarks and brands are registered property of their respective companies.

These Are Not Black Swans

7

We Knew the Storm Was Coming

Source: 1. Google Images search, public domain. All trademarks and brands are registered property of their respective companies.

And Yet, There Was Still Carnage

8

Halted Due to “Technical Issues”

Source: 1. Google Finance, chart. All trademarks and brands are registered property of their respective companies.

RADICAL PLATFORM PERFORMANCE

Cause and Effect? Performance Unleashed

10

Massive Capabilities. Small Investments.

Year GFLOP GB Mic

2007 $2.48 $48.00 $20,000,000.00

2012 $0.48 $0.92 $150,000.00

Decrease >5X >50X >130X

Cost Per:

Source: 1.The Aggregate, 2003-08-22, retrieved 2012-02-09 2. Adam Stevenson, Yann Le Du, and Mariem El Afrit. "^ Adam Stevenson, Yann Le Du, and Mariem El Afrit. "High-performance computing on gamer PCs." Ars Technica. March 31, 2011.." Ars Technica. March 31, 2011.

“Optimization” Is Relative Latency

Arb

Flow based execution arbitrage

Micro-market linear models

Machine Learning Models

Stochastic Models of order book dynamics

Statistical Arbitrage

GETCO

Breakwater

BATS

Rentec

Citadel

Automat

Registers

(1 Cycle)

Caches

(2-20 Cycles)

Memory

(100 Cycles)

Remote Memory

(10,000 Cycles)

Flash Drives

(100,000 Cycles)

Spinning Disks

(10,000,000 Cycles)

It is a blend of art, science...and strategy.

Put More Simply…

We Optimize Across A Range of Techniques

Latency vs Message Rate (Mpps)

0

5

10

15La

tenc

y (µ

sec)

Westmere 3.5 6.7 6.8 13.2

SandyBridge 2.5 2.6 2.9 3.2 3.8 4.7 7.7

1.0 1.5 2.0 2.5 3.0 3.5 4.0

Affordable Low(er) Latency

Moore’s Law Brings Lower Latency to the Masses More Easily Source: 1. Solarflare: http://www.solarflare.com All trademarks and brands are registered property of their respective companies.

http://www.solarflare.com/

The Latency Spectrum LA

TEN

CY

– LO

WER

IS B

ETTE

R

Parallelism; # of Threads

The Latency Spectrum LA

TEN

CY

– LO

WER

IS B

ETTE

R

PARRALELISM; # of THRAEDS

Apps that are truly single threaded and acutely latency sensitive represent an extremely small footprint of code.

Our Collective Challenge

16

const str newNormal = “parallel”;

0

50

100

150

200

250

300

350

400

450

2007 2008 2009 2010 2011 2012

Single Core Dual Core Triple Core Quad Core

Hex Core Octal Core 12 Core 16 Core

Units = M Source: PC Semiconductor Market Briefing: Re-Architecting the PC and the Migration of Value, June 2008

In 2012 all PCs & Servers shipping are Multi-Core, Multi-Threaded

GHz Era Multi-core Era

TIME PER

FOR

MA

NC

E

Multi-core Needs Parallel Applications

Perfo

rm

an

ce /

core

Core

Traditionally:

Improved performance with upcoming Intel® micro architectures

Next:

Leapfrog performance with wide vectorization, ISA extensions.

Intel Xeon 56xx Series • Intel® SSE4

• Intel AES-NI^-Cryptographic Acceleration

• Memory latency, BW

• Fast Unaligned support

Ivy Bridge

• Half float support

• Random Number Generator

Intel Xeon E5 Series

• Intel® AVX

• Up to 2X FLOP Output

• 3-Operand instructions

^AES-NI - Advanced Encryption Standard New Instruction

AVX – Advanced Vector Extensions

Future Extensions

• Hardware FMA

• Memory Latency/BW

• Many Other Features

Setting The Pace: Intel x86 ISA

Intel Architecture Goes Well Beyond “Speeds and Feeds”

Intel Xeon E5 - AVX Instructions

18

AVX Can Provide >1.8X Improvement to SSE Code

Newest in a long line of processor instruction innovations

Increases floating point operations per clock up to 2X1 for technical, financial, scientific & content creation applications

Performance to solve your most complex problems

Source: 1. Intel Corporation: http://www.intel.com All trademarks and brands are registered property of their respective companies.

http://www.intel.com/

Intel Fused Multiply Add (FMA) • Computes (ab)c with no intermediate rounding.

– Can speed up many common floating-point computations

• Get twice as many arithmetic ops per instruction

• Circuitry does not need to normalize intermediate result.

– Accurate computation of small residuals

• Varieties supported

types: float, double

widths: scalar, 128-bit vector, 256-bit vector

negations: ab+c, (ab)+c, abc, (ab)c,

19

And vector versions with alternating signs:

akbkck on even indices akbk+ck on odd indices

akbk+ck on even indices akbkck on odd indices

DEALING WITH DATA… BEYOND THE TAPE

A Cambrian Explosion… of Data

21

Photos uploaded to Facebook

Videos watched

Tweets sent

100M

2B

250M

7B

50M 230M

New Types of Data. New Types of Trading Opportunities. Source: 1. Public sources, Company Metrics from public filings: http://finance.google.com/(symbol) All trademarks and brands are registered property of their respective companies.

http://finance.google.com/(symbol)

Creating New Data Challenges

22

Intelligent Personalized Manageable

New Technologies and Methods are Required

A New Data Strategy

23

A New Data Strategy

24 Source: All trademarks and brands are registered property of their respective companies.

LOWER COST MODEL LEADS TO NEW DISRUPTIONS… AND COMPETITORS

Alpha From Tweets

26

Unstructured Graph Data

>2X Better Returns than Major Indexes During Test Period

Source: Correlating Financial Time Series with Micro Blogging Activity, Eduardo J. Ruiz, Vagelis Hristidis, Carlos Castillo, Aristedes Gionis, Alejandor Jaimes, February 8 -2012, Seattle Washingon, US – Copyright 2012

Analysis (And Execution) In the Cloud

27

Much Higher Alpha. Much Lower Costs.

Cloud Architectures

8 People, 4 Trades

>$X00M Gain… All from Cloud


Fully Optimizing For Success

28

Much Higher Alpha. Much Lower Costs.

Optimization Choices

Perf

orm

ance

Incr

ease


How Long Until You’re Optimized?

29

New Hardware, Cores, Sockets

Instructions

New Bios Settings

New OS / Java, Compilers

New Network New Tools –

1588., thread checker…

Optimized Application

Days? Months? Years?

If time is money, how much of both are you losing?

Start today. Start with Intel.

One Last Plug

30

Join in the discussion!


Thank you.

31 Source: All trademarks and brands are registered property of their respective companies.

Bending The Curve

Technology

Transcript of Bending The Curve