server channel training

Revision - 01

Copyright © 2008, Intel Corporation. *Other names and brands may be claimed as the property of others.

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.

• 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, visit Intel Performance Benchmark Limitations

• Intel does not control or audit the design or implementation of third party benchmarks or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmarks are reported and confirm whether the referenced benchmarks are accurate and reflect performance of systems available for purchase.

• Intel processor numbers are not a measure of performance. Processor numbers differentiate features within each processor family, not across different processor families. See http://www.intel.com/products/processor_number/ for details.

• 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.

• Intel Virtualization Technology requires a computer system with a processor, chipset, BIOS, virtual machine monitor (VMM) and applications enabled for virtualization technology. Functionality, performance or other virtualization technology benefits will vary depending on hardware and software configurations. Virtualization technology-enabled BIOS and VMM applications are currently in development.

• 64-bit computing on Intel architecture requires a computer system with a processor, chipset, BIOS, operating system, device drivers and applications enabled for Intel® 64 architecture. Performance will vary depending on your hardware and software configurations. Consult with your system vendor for more information.

• Intel, Intel Xeon, Intel Core microarchitecture, and the Intel logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries.

This slide MUST be used with any slides removed from this presentation

Quad-Core TechnologyUnprecedented Performance

Mainstream Value

(name)February 29, 2008

Revision - 01


Agenda

• Selling to Enterprise –Benefits of 45nm Quad-Core Technology

–Benefits of Data Center Refresh

• Selling to Small, Medium Business–2S Xeon Servers

–1S Xeon Servers

–1S Desktop Based Servers

• Selling Low Power Products

• Understanding Virtualization

© 2007 IDC

24.8%

21.4%

18.3%17.3%

10.2%

6.8%

0%

5%

10%

15%

20%

25%

30%

Reduce costs Increaserevenue

Increase marketshare

The Role of ITShifting Business GoalsThe Role of ITShifting Business Goals

Q. Prioritize the following business goals as they relate to your organization by allocating 100 points among them. The more points you allocate, the more important the business goal.

n = 225

Source: IDC’s Datacenter Trends Survey, January 2007

Reduce costs

Increase customer

satisfaction

Increase revenue

Improve quality/

accuracy

Increase market share

Speed time to market

Focus is again on growth– Top and bottom line goals are

usually at odds with each other, particularly increasing customer satisfaction

Still seeking temporary competitive advantages

Technology plays a more important role in future business initiatives

– New products

– New markets

– New constituents

– Regulatory requirements

Focus on users AND transactions

Grow Business

Control Cost

© 2007 IDC

State of Infrastructure TodayState of Infrastructure Today

Server Sprawl 41 m physical servers by

2010 - 700% increase in 15 years

Ave UT <10% = $140 bn in excess server capacity - a 3-year supply

Power & Cooling

50c for every $1 spent on server HW

$29 bn in power and cooling industry wide

Operating Cost

$8 in maintenance for every $1 spent on new infrastructure

20-30 : 1 server-to-admin ratio

Space Crunch

Datacenter construction: $1,000 / sqft $2,400 / server $40,000 / rack

Revision - 01


Greater Energy Greater Energy EfficiencyEfficiency

New 45nm Processors

Greater Greater Performance at Performance at

Given Frequency Given Frequency andand Higher Higher FrequenciesFrequencies

2nd Generation Intel Quad-Core

Quad-Core Intel® Xeon® processors (5400 series and 3300 series)

Technology Rich

Compelling IT Benefits

also available Dual-Core Intel® Xeon® processor 5200 series

New SSE4 New SSE4 InstructionsInstructions

Larger CachesLarger Caches

45 nm High-k 45 nm High-k Process TechnologyProcess Technology

Enhanced Enhanced Intel Core Intel Core

MicroarchitectureMicroarchitecture

Revision - 01


The 45nm Intel Advantage 45nm Quad-Core Intel® Xeon® Processor 5400 series Provide Key Benefits Over Prior Generation Quad-Core Processors

Energy Efficient

Virtualization

• up to 20% (mainstream servers)3

• up to 30%1 (most), 50%2 (optimized) HPC

• Faster Processors, Larger caches

• ~5x single core, ~2x dual core

Performance

• up to 38% better Perf/Watt4

• 16W idle power reduces system power

• Leadership SPECpower* results

• 80% energy savings & 2yr ROI on refresh

• Unique choices of server innovation

• Stable Mainstream: 5 CPUs 1 platform

• 1600MHz FSB & PCIe Gen2 for HPC/WS

• Cost/Power Optimized Entry Server

• Highest VMmark* score with 65nm Quad

• More native performance with 45nm

• Holistic approach to HW assist technology

• Intel® VT FlexMigration assist enables expanding compatibility pools

Innovation

1 (up to 30%) Source: Intel measured results Oct 2, 2007 for SPECfp_rate_base2006* comparing Xeon X5472 (3.0GHz/12M L2 Cache/1600 MHz FSB, 5400 Chipset (pre-production platform) to Xeon X5365 (3.00GHz/8M L2 Cache/1333 MHz FSB). 2 (up to 50%) 2 Source: Intel measured results Oct 2, 2007 for Financial Services (Black-Scholes application) running 8 threads with Red Hat EL4-U4 64-bit; 2.6.9-42.ELsmp comparing Quad-Core Intel Xeon 5300 Series based platform X5365 (3.00GHz/8M L2 Cache/1333 MHz FSB) to pre-production Quad-Core Intel Xeon 5400 Series based platform X5472 (3.0GHz/12M L2 Cache/1600 MHz FSB, 5400 Chipset (pre-production platform) 3 (up to 20%) Source: Published/measured results on SPECjbb2005* - Nov 12, 07 running Microsoft* Windows* Server 2003, SP1, x64 Edition with JVM: BEA JRockit* 5.0. Comparing pre production 2x Quad-Core Intel Xeon X5460 (3.16GHz/1333MHz,12M L2 cache) vs 2x Quad-Core Intel Xeon X5365 (3.00GHz/1333MHz/8M L2 cache) supporting 16GB system memory (FB DDR2-667 MHz). 4 (38% Perf/Watt) Published/measured results on SPECjbb2005* - Oct 2, 07. Comparison is between Xeon 5450 (3.00GHz/1333MHz/80W) and Xeon 5335 (2.33GHz/1333MHz/80W) Perf/Watt calculated by dividing the Performance by measured system power during steady state window. Any difference in system hardware or software design or configuration may affect actual performance

Revision - 01


Intel® Xeon® Server Product Family

ScalableScalableEnterpriseEnterprise

4p – 32p4p – 32p

IntelIntel®® Xeon Xeon®® 7300 7300(Intel® 7300 chipset)

High Performance High Performance ComputingComputing



2p2p

Mainstream Mainstream EnterpriseEnterprise

2p2p

Medium Medium Business Business

Small Small BusinessBusiness

2p2p

WorkstationsWorkstations

1p1p

Economical, Dependable Economical, Dependable 1-way Servers1-way Servers

Leading Performance, Leading Performance, Reliability & ScalabilityReliability & Scalability

Bandwidth Optimized for Bandwidth Optimized for High Performance AnalyticsHigh Performance Analytics

Maximize Performance Density Maximize Performance Density & Performance per Watt& Performance per Watt

IntelIntel®® Xeon Xeon®® 5400 5400(Intel® 5000 chipset)(Intel® 5100 chipset)


2p2p


Cost & Power Optimized Cost & Power Optimized Reliable Business ServersReliable Business Servers

Bandwidth Optimized for Bandwidth Optimized for Complex Analysis/VisualizationComplex Analysis/Visualization

Now with 45nm Quad-Core

Revision - 01


Boost IT Responsiveness with Compelling Performance Gains

0

2

4

6

8

10

12

14

2002 2003 2004 2005 2006 2007 2008

SP ECint_rate*_base2000

SP ECfp_rate*_base2000

2P TP C-C*

SAP -SD* 2 Tier

Linpack

Rela

tive P

erf

orm

an

ce

Up to

12x

Up to

9x

Up to

7x

Up to

6x

45nm delivers up to 20%more performance

over 65nm quad-core

Quad Core65nm 45nm

Quad-Core Benefits A Broad Range of Workloads Quad-Core Benefits A Broad Range of Workloads

Intel Multi Core Intel Single Core

Dual Core

Up to

2x

Published results. See backup for details. 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.

Revision - 01


20082008 5.1M bops5.1M bops 1 rack1 rack 17 Servers17 Servers 40 sq ft40 sq ft 6 kW6 kW

20042004 5.1M bops5.1M bops 6 racks6 racks 126 Servers126 Servers 240 sq ft240 sq ft 48 kW48 kW

Data Center Refresh

BOTTOM LINE BENEFITS for IT (< 2 year ROI4)

Estimated Estimated Energy SavingsEnergy Savings

$53K$53KSAVINGSSAVINGS33

AnnualAnnualEnergy CostsEnergy Costs

87%87%REDUCTIONREDUCTION22

Floor SpaceFloor Space

83%83%REDUCTIONREDUCTION11

BUSINESS REQUIREMENT: Deliver 5.1 million business operations per second

using 45nm Quad-Core Intel® Xeon® processor 5400

Source: Intel January 16, 2008. Performance comparison using SPECjbb2005 bops (business operations per second) between (2004) 2 socket single core Intel Xeon processor (3.6GHz) and (2008) 2 socket Quad-Core Intel Xeon processor E5450 (3.0GHz, 80W, 1333MHz) measured 8/22/07. 1 Floor space based foot on sq ft reduction. 2 Energy costs based on total solution rack power. 3 Energy Savings based on an electric rate of $0.10/kWh assuming 33% average server load. 4 ROI is calculated based on cost of new servers ($6,264 per server pricing based on HP DL 380G5 (32GB RAM) as of Nov 17, 2007. source www.hp.com) divided by energy savings per year. Actual performance results and savings may vary depending on configuration. See backup for details.

Revision - 01


Energy Efficient LeadershipSPECpower*

Rank Sponsor Score Platform Processors

1 HP 778 DL180 G5 2x Intel® Xeon® E5450


3 FSC 690 RX300 S4 2x Intel® Xeon® E5440

4 Dell 682 PE 2950 2x Intel® Xeon® E5440

5 FSC 667 TX150 S6 1x Intel® Xeon® X3220


7 HP 551 DL580 G5 4x Intel® Xeon® L7345

8 Intel 468 SM 6025B 2x Intel® Xeon® L5335

9 FSC 454 TX120 1x Intel® Xeon® 3070

10 FSC 446 RX300 S3 2x Intel® Xeon® L5335

11 FSC 356 TX150 S5 1x Intel® Xeon® 3070

12 Intel 338 SM 6025B 2x Intel® Xeon® 5160

13 AMD 203 CX2266-N2 2x Opteron* 2216 HE

14 Intel 87.4 White Box 2x Xeon 3.6/1M

SPECpower* first SPECpower* first realreal power power benchmarkbenchmark

Thermal Design

Power or Power Rating

Design Spec

Active Power

Measured under load

Idle Power

Measured w/o load

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, visit http://www.intel.com/performance/resources/limits.htm or call (U.S.) 1-800-628-8686 or 1-916-356-3104. Copyright © 2007, Intel Corporation. * Other names and brands may be claimed as the property of others.

Source: Public SPECpower results as of Jan. 31 2008

Created by a consortium of companies including AMD*, Dell*, Fujitsu Siemens*, HP*, IBM*, Intel, Sun* plus UC Berkeley*, Virginia Tech* and Lawrence Berkeley Labs*

45nm

45nm

45nm

45nm

45nm

Revision - 01


Competitive Brochure

Compare Quad-Core

Intel vs AMD*

Link to Brochure

Revision - 01


Relative SPECint_rate_base2006* Performance (2S DP)

0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 2.2 2.4 2.6 2.8 3 3.2 3.4 3.6 3.8

Opteron 2210, 1.80GHz, $174

Opteron 2212, 2.00GHz, $174

Opteron 2216, 2.40GHz, $255

Xeon X5205, 1.86GHz, $177

Opteron 2218, 2.60GHz, $377

Opteron 2220, 2.80GHz, $523

Opteron 2222SE, 3.00GHz, $698

Opteron 2224SE, 3.20GHz, $873

Opteron 2347, 1.90GHz, $316

Opteron 2350, 2.00GHz, $389

Xeon E5405, 2.00GHz, $209

Opteron 2360SE, 2.5GHz, TBD

Xeon E5410, 2.33GHz, $256

Xeon E5420, 2.50GHz, $366

Xeon E5430, 2.66GHz, $455

Xeon E5440, 2.83GHz, $690

Xeon E5450, 3.00GHz, $915

Xeon X5460, 3.16GHz, $1172

Xeon E5462, 2.80GHz, $797

Xeon E5472, 3.00GHz, $1022

AMD Quad-Core

Intel 45nm Dual-Core

AMD Dual-Core

$389$209

$177

$174

$377

Intel 45nm Quad-Core

future 2.5Ghz AMD QC

$256

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, visit Intel Performance Benchmark Limitations Copyright © 2007, Intel Corporation. Data Source: Published/measured/submitted results as of Nov 12, 2007.

Revision - 01


Quad-Core Intel® Xeon® Processor 5300 series Performance Comparison of X5365 & X5355 versus best published AMD Opteron*

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, visit Intel Performance Benchmark Limitations Copyright © 2007, Intel Corporation. * Other names and brands may be claimed as the property of others.

Xeon X5365 – Quad-Core Intel® Xeon® Processor X5365; Xeon X5355 – Quad-Core Intel® Xeon® Processor X5355; μ Dual-Core AMD Opteron* Model 2224SE; € Dual-Core AMD Opteron* Model 2222SE Φ Dual-Core AMD Opteron* Model 2220SE (2.80 GHz); Ψ Dual-Core AMD Opteron* Model 285 (2.60 GHz) δ Dual-Core AMD Opteron* Model 280 (2.40 GHz);

Data Source: Published or submitted results as of 09/25/07. Details in backup. Relative Performance. Higher is better

Quad-Core AMD Opteron* 2.0GHz results

Best available Dual-Core AMD Opteron* results used as baseline. Sept 25, 2007

21%26%

30%

47%

64%70%

76%80%

89% 90% 91% 92% 93% 93% 95% 95%

106% 109% 112%

131%

142%

0%

25%

50%

75%

100%

125%

150%

SP

EC

fp*_

rate

_b

ase

20

06

μ

SP

EC

fp*_

rate

20

06

μ

SP

EC

We

b*2

00

5€

We

b B

en

ch*Φ

Lig

htW

ave

3D

* R

en

de

rin

gtim

eΨ

Cin

eb

en

ch*

x C

PU

Ψ

SP

EC

int*

_ra

te_

ba

se2

00

6μ

TP

C-C

*Φ

Bla

ckS

cho

les*

Φ

SP

EC

ap

c* 3

ds

Ma

x* 8

Ψ

SP

EC

int*

_ra

te2

00

6μ

SA

P-S

D*

2-T

ierΦ

Flu

en

t* 6

.3.x

δ

PO

V-R

ay

3.7

B1

5 x

64

Ψ

Mo

nte

Ca

rlo

Pro

fit s

im*

-S

ecΨ

vCo

nso

lida

te-V

Mw

are

ES

XΦ

TP

C-H

10

0G

B€

Su

nG

ard

*A

CR

Φ

Lo

tus

Do

min

o*

7.0

Ψ

SP

EC

jbb

*20

05

μ

SP

EC

int*

_ra

te_

ba

se2

00

0Φ

Lin

pa

ck*Φ

TPC-C 80%

TPC-C 80%

Integer

91% (DC)

Integer

91% (DC)

VMware ESX 95%VMware ESX 95%

Java 131%

Java 131%

Top500

Linpack

227%

Top500

Linpack

227%

SPECfp Rate -

15% (QC)

SPECfp Rate -

15% (QC)Integer

31% (QC)

Integer 31% (QC)

Mainstream General Purpose 2P Servers

Rela

tive P

erf

orm

ance

Revision - 01


2 Y

EA

RS

2 Y

EA

RS

45nm

32nm

Shrink/DerivativeWestmere

New MicroarchitectureGesher

65nm

2 Y

EA

RS Shrink/Derivative

Presler · Yonah · Dempsey

New MicroarchitectureIntel® Core™ Microarchitecture

Shrink/Derivative Penryn Family

New MicroarchitectureNehalem

Intel® Micro-processor Design Model Tick-Tock

Quad-CoreIntel® Xeon®

processor 5400 / 3300 series

Dual-Core Intel® Xeon®

processor 5200 / 3100 series

featuring the industry’s first 45nm, high-k process technology

Revision - 01


What’s Next for Xeon-EP?

New 45nm Intel® MicroarchitectureNew Platform Design

Next Generation MemoryNext Generation I/O

Improved VT technology

Extending Today’s Leadership

Introduction Q4’08

Volume Ramp 1H’09

•More Application Performance

• Improved Energy Efficiency•End to End HW assist (virtualization)

•Stable IT Image– Software Compatible

– Live migration compatible with today’s dual and quad-core Intel® Core™ microarchitecture products using enabled virtualization software

Key Technologies

IT Benefits

PCI PCI ExpressExpress

**Gen 1, 2Gen 1, 2

I/OI/OHubHub

ICHICH

DMIDMI

Nehalem Nehalem

Functional system

demonstrated Sept 2007 IDF

Revision - 01


65nm Intel® Xeon® Processor 3000/X3200 Series

E3110 3.0/1333/6

X33602.83/1333/12

X3350 2.66/1333/12

Dual Core 3000 Series (codename “Conroe”) (65W)

Quad Core 3200 Series (codename “Kentsfield”) (95W)

30853.0/1333/4

30502.13/1066/2

X3230 2.66/1066/8

X32202.4/1066/8

30702.66/1066/4

30602.4/1066/4

X32102.13/1066/8

30752.66/1333/4

30401.86/1066/2

30652.33/1333/4

45nm Intel® Xeon® Processor X3300 Series

Dual-Core 3100 Series (codename “Wolfdale”) (65W)

Quad-Core 3300 Series (codename “Yorkfield”) (95W)

1066 FSB Dual-Core 1066 FSB Quad-Core 1333 FSB Dual-Core

X3320 2.5/1333/6

NEW!NEW!

Intel® Xeon® 3000 Sequence ProcessorsFor Intel® 3210 Chipset and Intel® 3200 Chipset-based serversIntel® Xeon® 3000 Sequence ProcessorsFor Intel® 3210 Chipset and Intel® 3200 Chipset-based servers

Intel may make changes to specifications and product descriptions at any time, without notice.

Model #GHz/FSB/cache

Revision - 01


45nm Performance Scaling for entry server solutions

45nm Benefits

Technology

Faster CPUs

More Cache

Performance

+ 20% vs 65nm

5x Desktop-based

Source: February 11, 2007. Published by Intel at http://www.intel.com/performance/server/entry_level/java.htm

5x existing desktop based

servers

+20%

45nm Quad-Core Intel® Xeon® Processor 3300 series

Revision - 01


Selling to Small, Medium Business

businessServer Technology enables your business

to support more users and more customers

with faster performance, higher reliability and security

Revision - 01


Small, Medium Business Needs

Vision: Improve Productivity and Capability While Reducing Costs

Business Uptime, Security and Virus Protection

Data Security & Reliability Proven Technology

Support Customers,Stay Ahead of Competitors

Application Performance

Contain Costs, Do More with Less

Price/PerformanceEnergy Efficiency

Low IT Headcount, Investment Security

Easy to Manage/Maintain Headroom, Flexibility

Business Need Technology Requirements

Revision - 01


What Type of Server?What Type of Server?

Three Typical Categories of Small Business ServersThree Typical Categories of Small Business Servers

1 Socket - Desktop Derived1 Socket - Desktop Derived

1 Socket - Server Validated1 Socket - Server Validated

2 Socket - Server Validated2 Socket - Server Validated

or or

Lowest CostGood for Basic NeedsProven Intel Technology

The Right Sized Entry ServerLow Cost Validated with Server ApplicationsImproved Reliability and I/O

Highest Application PerformanceBuilt with Business Growth in mind Enhanced Data Protection FeaturesExpandable Memory, I/O and Storage

Revision - 01


Server Usages Low Cost Entry Employee File Sharing • • •

Print Serving • • •Centralized Data Backup • • •

Central Firewall/Anti-Virus • • •FTP (Internet File Sharing) • • •

Groupware • • •Web Server • •

E-mail server • •Remote Access /VPN • •

Unified messaging • •Shared Accounting Apps • •

EDI •Customer Relationship Mgmt •

Remote Management •Database Server •

Enterprise •Business Analytics •

Virtualization •Intel® Pentium® D processor or Intel®

Core™ 2 Duo

Intel® Xeon® processor

3000 Sequence

Dual Intel® Xeon® processor

5000 sequence

Performance

Intel Small Business Server Solutions Intel Small Business Server Solutions

or

Revision - 01


Overcoming Customer Objections: Lack of Need

Proof Points:

Validation

• Intel spends 10,000+ man-hours validating every piece of an Intel® server rack

Data integrity

• ECC memory can detect and correct single-bit errors

• RAID subsystems and tape backups can help prevent data loss

Security

• Physical intrusion detection availableAvailability

• Hardware health monitoring and remote access and diagnostics

• Available redundant LAN, power, cooling, and storage capacity

“ My PC (or DTOS, or desktop-derived server) has been very reliable.”

Response:

As the volume of data you deal with grows, your PC won’t allow you to keep up. A real server based on the Intel® Xeon® processor brings you the peace of mind of:

Server validated components

Integrity and security of your data

Built-in reliability features

Revision - 01


Case Study: Lampwork, Etc*

After upgrading to a quad-core based server, Database queries and backup speeds have tripled giving Lampwork plenty of performance and growing room.

“We’re extremely excited about the growth possibilities that our new server gives us”

Task Performance Comparison 1

Username Change System Propagation

Old Server

New Server

Forums page response time(30 users online)

Old Server

New Server

Homepage response time during backup

Old Server

New Server

Database Back-up time (30 users)

Old Server

New Server

157s157s

3s3s

1.4s1.4s

4.2s4.2s

57ms57ms

259s259s

15s15s

Intel® Pentium® 4 521Intel® Pentium® 4 521 Quad-Core Intel® Xeon® Processor X3220Quad-Core Intel® Xeon® Processor X3220

Lampwork Etc* is a web-based business for people who make bead jewelry

Site UnresponsiveSite Unresponsive

“We knew that we’d hit the wall with our desktop-based server; it couldn’t bear the extra load. Our computer was suddenly a barrier to our growth, rather than a business enabler”Jason Harper, Lampwork Etc.

1 Performance results based onLampwork, Etc* internal application testing. Actual performance results may vary.

Read the complete case study at http://www.intel.com/business/casestudies/lampwork.pdf

Revision - 01


Overcoming Customer Objections Cost Barriers

Proof Points:

Improved reliability• The cost of downtime can quickly negate any

perceived savings from using a desktop• Using server management, you (or your service

provider) can proactively diagnose potential problems, helping you avoid costly downtime

Reduced maintenance costs • Intel spends 10,000+ man-hours validating

every piece of an Intel® server stackBetter ROI• Headroom enables you to support

more users• Enhanced performance enables you to

maximize the return on your software investment

Response:

If you include all the components required of a server in a desktop system, it will probably be more expensive and not support the full performance of the additional components. You also have to factor in the cost of business downtime as a result of lower reliability:

Unplanned IT administrative costs of data recovery (or worse, data loss)

Lost employee productivity

“ Servers cost too much.”

Revision - 01


First Server: Intel® Xeon®

1 Source: Benchmark conducted by Principle Technologies, Oct. 2007. Average performance of the three test servers across the three benchmarks relative to the Dual-Core Intel® Pentium® Processor E2100 based server. Higher numbers are better.

Quad-Core Intel® Xeon® Processor X3220 Based Server

Performance Benchmark for

Web, E-mail, and Database Workload1

100

199

Dual-Core Intel® Pentium® Processor E2100 Based Server

Desktop Based

Server Based

Almost double the

performance

For only $256

Lower maintenance costsMore Performance

Easy maintenance and repair

• Redundant components

• Remote diagnosis

• Easy access to components

• Validated for Server Environments

Get the work done faster with high confidence

Revision - 01


Feature

Intel Server Technology Matrix

Quad-Core Processing

• Provides FOUR physical compute engines for ultimate server performance

Memory Upgradeability

Key component if needing to upgrade OS or new software. Higher = better 8G 8G

Up to 64G

Low Cost Typical system hardware costs less than $1000

Server Validation

Platforms are validated to operate with Server operating systems such as Windows* Small Business Server and Linux

ReliabilityOffers server level reliability features such

as RAID and ECC Memory standard

Standard Server Feature Set

PCI Express x4 or x8, Dual Gigabit Ethernet

Extended Server Feature Set

2 CPUs, added Memory redundancy, hot swap and data protection (FBD), Accelerated I/O Technology (Intel I/OAT)

Source: Intel. As of July 11, 2007. All products, dates, and figures specified are subject to change without notice.

BenefitDesktop Derived

Xeon 5000 Platform

Xeon 3000 Platform

Revision - 01


Pentium® D (1 Processor, 1 core )

Xeon® 3000 (1 Processor, 2 cores)

Xeon® 3200 (1 Processor, 4 cores)

Xeon® 5300 (2 Processors, 8 cores)

Quad-Core Intel® Xeon® ProcessorsAn outstanding choice for your small business

Quad-Core Intel® Xeon® ProcessorsAn outstanding choice for your small business

51,476

112,092

210,599

31,138

Application Performance

(java performance)

Operations per secondSPECjbb2005*

51

101

200

38

Data Throughput(Integer performance)

Jobs per hourSPECint*rate_base2000

8102

10971

18160

2151

Online Web Hosting

(web experience)

Published results as of April 10, 2007 on www.spec.org using SPECjbb2005* (java), SPECint*_rate_base2000 (integer), SPECweb2005* (web). For detailed performance and configurations consult www.intel.com/performance. 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, visit http://www.intel.com/performance/resources/limits.htm or call (U.S.) 1-800-628-8686 or 1-916-356-3104.

Performance & CostPerformance & Cost

Relative Web performance (Higher is better)SPECweb2005*

Single CoreSingle CPU

Dual CoreSingle CPU

Quad CoreSingle CPU

Quad CoreTwo CPU

$300-500

$500-1,000

$1,000+

Typical System Hardware Cost2

$500-1,000

2 system cost estimates are general ranges as of July 2007.

Revision - 01


TodayToday 1.7M bops1.7M bops 6 Servers 6 Servers 1 rack1 rack 40 sq ft40 sq ft 2.1 kW2.1 kW

Source: Intel. Performance comparison using SPECjbb2005 bops (business operations per second)

20042004 1.7M bops1.7M bops 42 Servers42 Servers 2 racks2 racks 80 sq ft80 sq ft 16 kW16 kW

Server Refresh: A 2year ROI

BOTTOM LINE BENEFITS for IT

Estimated Estimated Energy SavingsEnergy Savings

$18K$18KSAVINGSSAVINGS

AnnualAnnualEnergy CostsEnergy Costs

86%86%REDUCTIONREDUCTION

Floor SpaceFloor Space

50%50%REDUCTIONREDUCTION

BUSINESS REQUIREMENT: Deliver 1.7 million business operations per second

using 45nm Quad-Core Intel® Xeon® processor 5400

Room to

Grow

Revision - 01


Selling Lower Power

Power Optimization Techniques to …

– Put more servers in the rack

– Boost performance per rack

– Optimize performance per square foot

In existing rack infrastructure

Revision - 01


New 2P Entry Business Server

Optimized for businesses seeking enterprise class

performance and quality on a constrained budget

New Platform Q4’07• High performance multi-core CPUs

• Excellent investment protection• Quad-Core and Dual-Core capable

• Power efficient Intel® Core™ µarch

• 6 slots of lower power DDR2 Memory

• Cost optimized chipset and I/O controller

• 30 lanes of configurable PCIe

6 slots24-32 GB max

DDR2DDR2

DDR2DDR2

ICH-9RICH-9RI/O BridgeI/O Bridge

Quad-Core & Dual-Core5400 Series5300 Series5200 Series5100 Series2x2x

1333 MHz1333 MHz1066 MHz1066 MHz

Small, Medium Business Server

High Performance, Low Cost, Energy Efficient

boost business without breaking the budget

Configurable Configurable PCI Express*PCI Express*

30 lanes PCIe

Intel 5100

NEW Design November 2007

Revision - 01


Optimizing Power with Intel

• Use newest 45nm Intel Xeon processors3

– Lowers idle power consumption by up to 2/3 relative to

65nm processors

• Consider Low Voltage Processors2

– Another 30W lower power per CPU (L53xx, L54xx SKUs)– 40-50W lower platform power for 2 socket configurations– Same performance

• Use Low Power Intel®

5100 chipset1

– When small memory capacity (<12GB) is sufficient … – 30-50W lower platform power– Similar platform performance

Source: Intel, November 27, 2007.1 Comparisons is between similarly configured Intel Xeon 2P servers using Intel 5100 chipset and the Intel 5000 chipset 2 Comparison is between using 50W quad-core Xeon 5300 series and 80W Quad-Core Xeon 5400 series3 Comparison is between Quad-core Intel Xeon 5400 series vs Quad-core Intel Xeon 5300 series

Revision - 01


Boosting Performance & Density (modeling SPECint_rate_base2006* and small memory footprint)

6kW Rack 6kw Rack 6kw Rack

~340W server ~300W server ~240W server

2 x 80W processorsXeon E5420 (2.5GHz)

Intel 5000 chipset12GB Memory / server

2 x 50W processorsXeon L5420 (2.5GHz)

Intel 5000 chipset12GB Memory/server

2 x 50W processors L5420 (2.5GHz)

Intel 5100 chipset12GB Memory/Server

17Servers

20Servers

24Servers

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, visit Intel Performance Benchmark Limitations. See backup for details

+15%

L54xx

L5320

L5310

50Wquad-core

L52xx

5148

40W dual-core

Baseline

+38%

L54xx and L52xx products scheduled for production release Q1’08

LV Xeon LV Xeon

+

Low Power

platform

Revision - 01


Understanding Virtualization

Revision - 01


Virtualization: Why?Different Operating

Systems

Multiple Applications

Secure Partitions

One Server

KEY: App = Application, Mgmt = Management, OS = Operating System, VM = Virtual Machine

Business Benefits of VirtualizationSmaller IT Environment, Lower Power Bills

Less Servers to manage and maintainEasier and Quicker to Add New ApplicationsWithout Sacrificing Application Performance

Advanced Business Computing … Proven Benefits

One Quad-Core Intel Xeon

processor 5000 sequence-based servers

Revision - 01


Does HW Matter? … Yes

Performance Headroom Quad-core and Intel® Core™ microarchitectureLarger servers support more memory and I/O

Hardware Assist Features 64 and 32bit OS on the same server

Reduced Virtualization Overhead (cpu, memory, i/o)Functionality: Live VM migration across CPU generations

Lower costs over time with OS integration

Consolidate with High Confidence Proven dependability with unique reliability features

Intel® Xeon® Processor are Ready When You Are 2S Quad-core Intel Xeon Servers Are Built For Virtualization

KEY: App = Application, Mgmt = Management, OS = Operating System, VM = Virtual Machine

One Quad-Core Intel Xeon

processor 5000 sequence-based servers

Revision - 01


Quad-Core Virtualization Benefits

30% lower TCO 1

60% more Performance

30% more Performance/$

20% more Performance/Watt

Intel IT testing shows that the Quad-Core Intel® Xeon® processor-based servers are an excellent platform for running typical IT workloads with hypervisor-based virtualization software

1IT@intel study: http://www.intel.com/it/pdf/boosting-server-virtualization-performance.pdf

Revision - 01


Why Virtualize with Intel Inside

Top virtualization results1

for VMware* Vmark*

1 Source: VMware as of Dec 11, 2007. See: http://www.vmware.com/products/vmmark/results.html for details. 2 Intel Virtualization Technology requires a computer system with a processor, chipset, BIOS, virtual machine monitor (VMM) and applications enabled for virtualization technology. Functionality, performance or other virtualization technology benefits will vary depending on hardware and software configurations. Virtualization technology-enabled BIOS and VMM applications are currently in development.

Intel does not control or audit the design or implementation of third party benchmarks or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmarks are reported and confirm whether the referenced benchmarks are accurate and reflect performance of systems available for purchase.

Performance Hardware Assist

Intel® Virtualization Technology (VT) is a set of hardware enhancements that eclipse traditional software-only based virtualization

Processor:

Intel® VT FlexMigration2: Move virtual machines between servers without service interruption

Intel® VT FlexPriority2: improves interrupt handling in virtual environment, boosting performance by up to 27% for 32bit guest operating systems

Chipset & IO:

Intel® I/O Acceleration Technology2: Offloads routine I/O tasks from processor to I/O devices freeing up 35% more CPU cycles for virtualization

Intel® VT for Connectivity2: Specialized networking devices sort data through multiple queues, delivering more efficient networking

45nm

Revision - 01


Intel-based ServersIntel-based Servers

Built for Performance and Energy EfficiencyBuilt for Performance and Energy Efficiency

Learn more: www.intel.com/quadcoreserver

Thank You for

Considering Intel

Revision - 01


Backup

Revision - 01


Stable and Reliable

Energy EfficientPerformance

Built for Virtualization

Addressing IT Challenges Head OnAddressing IT Challenges Head OnSince 2006, millions of mainstream 65nm Quad-Core Intel® Xeon® processors have helped IT Reduce Costs, Improve Business Continuity & Responsiveness

1 Intel internal measurement as of Oct 9, 2006 using SPECint_rate_base2000*. Results compare Xeon E5345 (CTN 2.33/1333/80W) with measured Xeon 5160 (WC 3.00/1333/80W), published Dual-Core AMD Opteron* Model 2220SE (2.80 SktF) and published 64bit Intel Xeon Processor3.60GHz with 2ML2 cache (Irw 3.6/ LH800/110W). 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.

More Performance Less Space

More Technology Fewer Risks

•Dual + Quad: same server •Advanced memory, I/O •More Reliability Features

•1.5x over Xeon® Dual-Core1

•2.5x over other’s dual-core1

•4.5x over Xeon® single core1

•80W quad-core processors

•2x cores, cache, threads•Hardware-assisted tech•Broadest SW ecosystem

Doing More With Less

Revision - 01


Intel® Core™ Microarchitecture(introduced 2006)

Intel® Core™ Microarchitecture(introduced 2006)

Intel® Advanced Smart Cache Intel® Smart Memory Access

IntelIntel®® Intelligent IntelligentPower CapabilityPower Capability

Intel® Wide Dynamic Execution Engine

FASTER

SMARTER

EFFICIENT

4 instructions per cycle

2x cache size, sharedAdvanced pre-fetching

ultra fine grained power control

How do these features work?

Learn more at intel.comhttp://www.intel.com/technology/architecture/

coremicro/demo/demo.htm

Delivering a Performance and Power Advantage for IT

Revision - 01


Up to 20% Higher Performance

1

1.18 1.18 1.2 1.211.12

Baseline SPECfp*_rate2006

SPECint*_rate2006

SAP-SD* VMmark* SPEC*jbb2005

Quad-Core Intel Xeon

X5365

Platform of Choice Since mid’06 adds 45nm processors

Nov’ 07: More Performance - Stable Platform

General Purpose

Java Server

Scientific Comp.

ERP Server

Virtual- ization

Quad-Core Intel® Xeon® X5460

Data source: Published/Measured results as of Nov 12, 2007. See backup for detailsPerformance 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, visit Intel Performance Benchmark Limitations

Raising the Bar for PerformanceMainstream servers

Revision - 01


What is SPECpower*• SPECpower* is a new industry standard benchmark for measuring

energy efficiency of volume servers

• SPEC – The Standard Performance Evaluation Corporation released SPECpower_ssj2008* on December 11, 2007

• SPECpower_ssj2008* is the first industry standard server benchmark to measure power consumption in relation to performance across a “graduated” workload

• SPECpower benchmark report provides unprecedented data on platform power consumption across load level range

• Created by a consortium of companies

– AMD*, Dell*, Fujitsu Siemens*, HP*, IBM*, Intel, Sun* plus UC Berkeley*, Virginia Tech* and Lawrence Berkeley Labs*

SPECpower* Fills a Need For a Standard Method to Measure Power And Performance Together

Revision - 01


Why Virtualize? Why Virtualize?

Initial Advanced

Server ConsolidationTest and Development

Dynamic Load Balancing

Disaster Recovery

IT Usages

IT Benefits• Fewer Servers Lower IT Costs

• Software Flexibility Higher Utilization

• Less Power Easier Cooling

Real Question for IT is Why Not?

Revision - 01


4P+

2P

Intel® Core™ Microarchitecture

Enhanced Intel® Core™ Microarchitecture

Next Generation Intel Microarchitecture

1P

‘Dunnington’

Xeon 5400

Xeon 3300

Future

Future

Future

1Backward compatibility for live VM migration also exists with current dual-core Intel® Core™ microarchitecture products (Xeon 5100 and Xeon 3000) and forward compatibility with future dual and multi-core processors. Contact your preferred VMM vendor for support requirements.

Xeon 5300

Xeon 7300

Xeon 3200

Merom based Penryn based Nehalem based

2006-2007

2007-2008 2008-2010

µarch

Dual Core 5100

Dual Core 7200

Dual Core 3000

All future products, dates, and figures are preliminary and are subject to change without any notice.

Intel® VT FlexMigration assistIncreased VM Mobility with VMware

Fail-Over

Load Balance

Disaster Recovery

Server Maintenance

VM

VM

IT Investment Protection with future ESX* versions

Flexibility & Investment ProtectionIntel® VT FlexMigration assist and VMware*

Flexibility of Live VM Migrationusing ESX* solutions Today

Dual Core 5200

Dual Core 3100

Revision - 01


Online Tool

Provides Quick Estimate of IT Value

with Quad-Core

Native Performance Gains

and

IT Cost Savings with Virtualization

Quad-Core ROI Tool

www.intel.com/go/virtualization

Revision - 01


Why a Server? business

A server provides a dedicated, centralized, networked machine that provides services to local and or remote PCs – Print and internet connection sharing– Centralized file sharing– Secure back-up of client data– Database and other shared applications

Server Technology enables your business

to support more users and more customers

with faster performance, higher reliability and security

Revision - 01


Technology Built for Business

Intel Leading Technologies – Benefits for Small & Medium Business

Better storage and data protection•Intel® Execute Disable Bit•Intel® Matrix Storage Technology

More Performance for your $

•Multi-core processors starting sub $200 1 •Intel® Virtualization Technology•Advanced manufacturing, (65nm, 45nm)

Industry Leading performance

•New Intel® Core™ Microarchitecture•Intel Wide Dynamic execution•Intel® Advanced Smart Cache•Intel Smart Memory Access

Denser, quieter & lowerpower servers

•Intel® Intelligent Power Capability•50W low voltage quad-core processors•Energy-Efficient Performance

Increased Utilization & Flexibility

•Intel® Virtualization Technology•Intel®VT FlexMigration•Intel®64 Architecture

Highest Reliability, Ease of Use

•ECC error correcting memory•Genuine Intel Architecture•Fully Buffered DDR2 Memory

High Quality Platform

•Comprehensive Server Validation•Architectural Planning Security•Genuine Intel® Architecture

1 Source Intel. Subject to change without notice

Revision - 01


Servers with Intel® 5100 Chipset• Qualifying Accounts: Optimized for customers with …

– Small memory needs (12GB or lower)

– Older rack infrastructure (<6kW per rack)

– No near-term plans for virtualization (server utilization remains low)

• Ideal Users: Small, Medium Business and Emerging Markets

• Recommended Sales Approach (Enterprise)1. Lead with ‘Bensley’: Mainstream 2S Xeon servers with Intel 5000 chipset2. If power and rack density is THE objection … introduce Intel 5100 chipset

Advantages1 Tradeoffs1

Similar Server PerformanceLimited to 6 slots of memory (12GB with 2GB DIMMS)

~30-50W lower server power

Source: Intel, November 27, 2007. Comparisons is between similarly configured Intel Xeon 2P servers using Intel 5100 chipset and the Intel 5000 chipset

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, visit Intel Performance Benchmark Limitations. Copyright © 2007, Intel Corporation. * Other names and brands may be claimed as the property of others.

Revision - 01


69.2

109

63.3

99.9

198

251

Improved Performance per WattIntel® 5100 chipset for small system memory configurations

Lower Platform PowerLower Platform Power

Measured System Power in Watts during benchmark steady state

Lower is better

20% Lower Power

Intel® 5000

Chipset

Intel® 5100

Chipset

Intel® 5000

Chipset

Intel® 5100

Chipset

SPECfp*_rate_base 2006 SPECint*_rate_base 2006

Similar Server Performance Similar Server Performance

Intel® 5000

Chipset

Intel® 5100

Chipset

Technical Computing(floating point)

Both measurements feature two Quad-Core CPUs

(Xeon E5450) and 8GB memory (4x2GB memory)

General Purpose(integer throughput)

Higher is better

Data Source: Internal measurement as of 5 Nov 07. Actual performance may vary. See backup for details.


Similar Performance

8GB memory (4x2GB memory)

Revision - 01


Quad-Core PerformanceQuad-Core Performance 45nm, 245nm, 2ndnd Generation Quad-Core Intel Generation Quad-Core Intel®® Xeon Xeon®® Processor 5400 Series Processor 5400 Series

Technical Compute Servers

1

1.211.19#

Quad-Core IntelXeon X5365

3.0GHz

Quad-Core IntelXeon 5400

Series 3.16GHz


Series 3.16GHz

Mainstream Enterprise

Servers

1 Manufacturing Workflow: defined as concurrently running SPECapc*

SolidWorks* 2005 and Fluent* 6.3.26 L1/L2, 5x concurrently

SPECint*_rate

2006JAVA*

Source: Intel. Results on Quad-Core Intel Xeon 5400 series are Intel internal measured results, October 2, 2007. All other Quad-Core Intel Xeon results are based on measured/published results. Quad-Core Intel Xeon Processor X5365: www.spec.org, Current as of 9/17/2007. Details in backup.

Integer Java*

Floating Point

Bandwidth Intensive Applicationsgeneral purpose

SPECfp*_rate

2006

Extending Performance Leadership Extending Performance Leadership

+50%+50%+20%+20%1

1.34

1.23


3.0GHz


Series 3.20GHz


Series 3.20GHz

1

1.5

1.4


3.0GHz


Series 3.00GHz


Series 3.00GHz

Workstation &

Multitaskingconcurrent applications1

Design&

Analysis

+34%+34%

SPEC fp*_rate

2006

Floating Point

WeatherModeling

FinancialServices

up to up to up to

Black-Scholes*

National Weather Services

Application *

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, visit Intel Performance Benchmark Limitations

Revision - 01


19.8

30.5

42.2

47.5

I ntel® Xeon® Processor X3360I ntel® Xeon® Processor X3230I ntel® Xeon® Processor 3085I ntel® Pentium® D Processor 945

Compute-Intensive Throughput

Est. SPECfp*_rate_base2006

Data Source: Submitted or published results as of 7 Jan 2008. Actual performance may vary. See backup for details.

1.0

1.5

2.1

2.4

• 3.40 GHz• 2 MB

cache• 800 MHz

bus

• 3.00 GHz• 4 MB

cache• 1333

MHz bus

• 2.66 GHz• 8 MB

cache• 1066

MHz bus

• 2.83 GHz• 12 MB

cache• 1333

MHz bus

20.6

34.9

55.1

63.4

I ntel® Xeon® Processor X3360I ntel® Xeon® Processor X3230I ntel® Xeon® Processor 3085I ntel® Pentium® D Processor 945

Est. SPECint*_rate_base2006

1.0

1.7

2.7

3.0

Baseline

General PurposeTechnical Computing

Normalized as 1.0 to Pentium D 945 using Intel® 3010 Chipset-

based server. Higher is better

20

06

-20

07

20

07

-20

08


NormalizedNormalized NormalizedNormalized

Entry Server Platform Performance featuring Intel® Xeon® Processor 3300 Series on Intel® 3210

Chipset-based servers

New! 45nm Quad-Core

65nm Quad-Core

Dual-Core

Dual-Core

New! 45nm Quad-Core

65nm Quad-Core

Dual-Core

Dual-Core

Revision - 01


Backup - Data Center OptimizationSystem Configurations:2 Socket Intel® Xeon® Processor 3.6GHz (Nacona)8 x 1GB DDR2-400 DIMMsSystem Idle Power: 225WSystem Workload Power: 382WSPECjbb2005 bops performance: 40487

2 Socket Intel® Xeon® Processor E5450 (3.0GHz 80W HTN)8 x 2GB 667MHz FBDIMMsSystem Idle Power: 237WSystem Workload Power: 353WSPECjbb2005 bops performance: 297,984

Calculations:Floor space: 80sq. Foot to 40 sq. foot (50% reduction)Annual energy costs: 102.3kW to 13.8kW (87% reduction)Yearly Energy Savings: $17.7k ($20.5k - $2.76k) = $17.7kQty 42 2S Intel® Xeon® Processor 3.6GHz (Nacona) servers, at idle power (225W * 16 hours/day), at full load (382W * 8 hours/day), for one year at 0.10 kWh = $10.23k *

2 (HVAC factor) = $20.5k electricity cost per year.Compared toQty 6 2S Intel® Xeon® Processor E5450 (HTN 3.0GHz) servers, at idle power (237W * 16 hours/day), at full load (353W * 8 hours/day), for one year at 0.10 kWh = $1.38k

* 2 (HVAC factor) = $2.76k electricity cost per year.

ROI: Using HP DL 380G5 with 32GB RAM is $6,264 as of Nov 17, 2007 on hp.comCost of 6 servers is $37.6k divided by $17.7k/year energy savings, this is 2.12 year ROI

Software:BEA JRockit* (build P27.4.0-3-86647-1.6.0_02-20070801-1931-windows-x86_64, compiled mode

Revision - 01


Backup: Optimizing Performance / Rack

• Power– 340W per server: 2* 80W E5450 (3.0GHz), Intel® 5000 Chipset, 6* 2GB DIMMs

> Source: Intel internal measurements + 22W (11W per DIMM * 2). > Measured platform: Supermicro SYS-6025B-TR+V (EBI2SQC61), memory: 4* 2GB; DDR2-667

– 300W per server: 2* 80W E5450 (3.0GHz), Intel® 5100 Chipset, 6* 2GB DIMMs> Source: Public SPECPower measurements from HP DL180 G5 server

http://www.spec.org/power_ssj2008/results/res2008q1/power_ssj2008-20080116-00033.html as of Feb 27, 2008 plus 60W (2*30W TDP increase)

– 240W per server: 2* 50W L5420 (2.5GHz), Intel 5100 chipset, 6* 2GB DIMMs > Source: Public SPECPower measurements from HP DL180 G5 server

http://www.spec.org/power_ssj2008/results/res2008q1/power_ssj2008-20080116-00033.html as of Feb 27, 2008

• Performance– Performance based on SPECint_rate_base2006 public data as of Feb 27, 2008. – Benchmark score of 109 for E5450 and 77.3 for E5420 and L5420