This Intel® Xeon® Processor E5-2600 v3 Product Family Application Showcase focuses on Core Business Applicatoins software companies who have seen preformance increases with Intel products.
Unraveling Multimodality with Large Language Models.pdf
Intel® Xeon® Processor E5-2600 v3 Product Family Application Showcase - Core Business Applications
1. Intel Confidential — Do Not Forward
Intel® Xeon® Processor E5-2600 v3 Product Family
The Fast Lane to Better Performance†
Application Showcase:
Core Business Applications
November 2015
How fast do you want to go? † Better Performance is demonstrated through proof points in this presentation.
New
Proof
Points
2. The New
Foundation for
Software-Defined
Infrastructure
2
Common Architecture for
Compute, Storage and Network
Efficient, High Performance
Building Blocks
System Visibility for
Monitoring and Control
Intel® Xeon® Processor E5-2600 v3 Product Family
The Intel® Xeon® processor E5-2600 v3 product family is at
the heart of your agile, efficient data center, meeting your
needs for compute, storage, and network.
These energy-efficient marvels are designed to deliver the
best combination of performance, built-in capabilities, and
cost-effectiveness and are designed with technical
computing, cloud deployments, intelligent storage, design
automation and data analytics in mind.
Safeguard your data and infrastructure with foundational
technology for greater security. Break challenging data
management performance bottlenecks with 40 gigabit Intel®
Ethernet, DDR4 support and innovative storage features that
make Intel-powered platforms the smart choice for a
software defined world.
You only have to remember one thing: It’s what’s inside that
counts.
3. Up to Up to 6X increase in
performance with Intel® AVX25
Up to 3X increase in virtual
machine density6,7
Up to 3X increase in memory
bandwidth with DDR48
3
Delivering Leadership Performance
27 New World Records1
!
Up to 3X performance
improvement for efficient, high
performance building blocks3
Up to 6X throughput
with the latest Intel® DC SSD4
Up to 40GbE to improve
network virtualization
The World’s Most Energy Efficient Server2
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. For more information go to http://www.intel.com/performance Intel does not control or audit third-party benchmark data or the
web sites referenced in this document. You should visit the referenced web site and confirm whether referenced data are accurate.
*Other names and brands may be claimed as the property of others.
1. Twenty-seven performance world records based on two-socket configurations. Source as of September 8, 2014. Full details available at: http://www.intel.com/content/www/us/en/benchmarks/server/xeon-e5-2600-v3/xeon-e5-2600-v3-summary.html
2. Comparison based on SPECpower_ssj2008results published as of August 26, 2014. New configuration: Sugon I620-G20platform with two Intel Xeon Processor E5-2699v3, 8x16GBDDR4-2133 DR-RDIMM, Microsoft Windows Server 2008 R2, IBM J9 VM, 10,599overall ssj_ops/watt. Source
Other names and brands may be claimed as the property of others.
3. Source as of September 8, 2014. New configuration: Hewlett-Packard Company HP ProLiant ML350Gen9 platform with two Intel® Xeon® Processor E5-2699 v3, Oracle Java Standard Edition 8 update 11, 190,674SPECjbb2013-MultiJVM max-jOPS, 47,139SPECjbb2013-MultiJVM critical-jOPS. Source. Baseline: Cisco Systems Cisco UCS
C240 M3 platform with two Intel® Xeon® Processor E5-2697v2, Oracle Java Standard Edition 7 update 45, 63,079 SPECjbb2013-MultiJVMmax-jOPS , 23,797SPECjbb2013-MultiJVM critical-jOPS. Source.
4. Comparisons based by Intel on publicly available specification sheet data &/or white papers
5. Source as of August 2014on Linpack*. Supermicro* X8DTN+ with two Intel® Xeon® processor X5690, RHEL* 6.1, 12x4GB DDR3-1333, Intel® HT Technology disabled, Intel® Turbo Boost Technology enabled, SMP Linpack 10.3.5. Intel internal measurements TR#1236. Score: 159.36 GFlops. New configuration: Intel® Server System
R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3, Intel® HT Technology disabled, Intel® Turbo Boost Technology enabled, 8x16GB DDR4-2133, RHEL* 6.4, Intel® MKL 11.1.1. Intel internal measurements TR#3034. Score: 1,012 GFlops.
6. Intel does not control or audit the design or implementation of third party benchmark data or Web sites referenced in this document. Intel encourages all of its customers to visit the referenced Web sites or others where similar performance benchmark data are reported and confirm whether the referenced benchmark data are accurate
and reflect performance of systems available for purchase.
7. Source as of September 8, 2014. Baseline: Fujitsu PRIMENERGY RX300 S6 with two Intel® Xeon® Processor X5690, VMware ESXi 4.1 U1, VMmark* v2.1.1 score: 7.59 @ 7 tiles, source. New Configuration: Fujitsu PRIMERGY RX2540 M1platform with two Intel® Xeon® Processor E5-2699v3, VMware ESXi 5.5.0 U2, VMmark v2.5.2 score:
26.48@ 22 tiles, source. VMware® VMmark® is a product of VMware, Inc.
8. Source as of August 2014TR#3044 on STREAM (triad): Supermicro X8DTN+ platform with two Intel® Xeon® Processor X5680, 18x8GBDDR3-800 score: 26.5 GB/sec. New Configuration: Intel® Server System R2208WTTYSwith two Intel® Xeon® Processor E5-2699v3, 24x16GB DDR4-2133@ 1600MHz DR-RDIMM, score: 85.2 GB/sec.
4. Common Architecture Enhancements
1st DDR4 Memory
Implementation
Intel® Advanced
Vector Extensions 2.0
Intel® Data Direct I/O
Intel® Intelligent
Storage Acceleration
Library
Intel® QuickAssist
Technology
Intel® Data Plane
Development Kit v1.7
Compute Storage Network
4
10. 10
Nomura Research Institute, Ltd.*
Business Oriented Solution (BOS) – Real-time stock calculation;
Normalized TPC-H benchmark schema with Bill of Materials
“NRI* is excited about the Intel® Xeon® processor E5-2600 v3 which
significantly improves performance of our NUMA optimized applications. Not
only more CPU cores and faster memory I/O but also Cluster on Die (CoD)
really boosted performance to amazing levels compared to the previous
generation processors”
Yuzo Ishida – Senior Application Engineer, NRI (Architect of BOS)
Real-time Enterprise Architecture
that scales with NUMA optimizations
Core Business Applications
Intel® Xeon® Processor E5-2600 v3 Product Family
Real-time enterprise applications sharing single database schema
(building the arbitrary number of logical views on the fly from
horizontally scalable database nodes by “functional partitioning”)
Just like the higher CPU clock rate had directly contributed better
performance, more CPU cores with low latency between CPU and
memory improves the performance of BOS type applications without
touching source code.
http://www.nri.com/
0
1
2S Intel® Xeon® processor E5-2697 v2
(24 cores / 2 NUMA Nodes)
2S Intel® Xeon® processor E5-2699 v3
(36 cores / 4 NUMA nodes)
Product increased performance with Intel® Xeon®
processor E5-2699 v3 enabling Cluster on Die (CoD)
Up to
62%
Faster
N E W
1 - Testing conducted on MariaDB* 10.1 comparing Intel® Xeon® Processor E5-2697 v2 with Intel® Xeon® Processor E5-2697 v3. Testing done by Intel. For complete configuration details see slide 41.
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.
XPerformanceIncrease
11. 0
1
Intel® Xeon® processor E5-2697
v2
MariaDB Corporation*
MariaDB 10.1; typical R/O OLTP requests, 1M table size and 10M RW.
Multi-table scenario workload
“The new Intel® Xeon® processor E5 -2600 v3 continues the dual-socket
performance evolution. Powered by additional enhanced cores, it improves
performance for traditional databases by up to 1.39X1 vs. the previous
generation. Our customers will get faster responses to their complex queries,
and can reduce TCO at the same time – so they can realize even higher value
with MariaDB”
Michael Widenius - MariaDB Foundation CTO
Intel® Xeon® Processor E5-2600 v3 Product Family
11
www.mariadb.org
Robust, scalable, and reliable SQL server. Drop-in replacement for MySQL*.
Intel® Xeon® processor E5-2600 v3 with enhanced microarchitecture,
additional enhanced cores and larger cache give the performance boost
delivers a significant performance gain for traditional database
applications such as MariaDB.
Faster response to complex queries, lower TCO.
Faster response times and lower TCO on a popular,
open-source database; 1.39X speed-up!
Core Business Applications
1 - Testing conducted on MariaDB* 10.1 comparing Intel® Xeon® Processor E5-2697 v2 with Intel® Xeon® Processor E5-2697 v3. Testing done by Intel. For complete configuration details see slide 41.
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.
XPerformanceIncrease
MariaDB* 10.1 increased performance (18,961 to
26,437 transactions/second) with the Intel® Xeon®
processor E5-2697 v3
Up to 39%
faster!
INTEL CONFIDENTIAL – USE UNDER NDA ONLY *Other names and brands may be claimed as the property of others
N E W
12. 0
1
Intel® Xeon® processor E5-2697 v2
Intel® Xeon® processor E5-2697 v3
Oracle Corporation*
MySQL Cluster* 7.4 - flexAsynch benchmark
“We observe a 33% performance boost for MySQL Cluster* with the new
Intel® Xeon® processor E5-2697 v3). We achieved 156M reads per second vs.
117M running the flexAsynch benchmark on a 24 node cluster vs. the
previous processor generation. Additionally, we achieved a new record of
200 Million Queries Per Second (QPS), or over 12 Billion QPM, with MySQL
Cluster! The record was established using a cluster running 32 Intel Xeon E5-
2600 v3-based nodes.”
Tomas Ulin – VP, MySQL Engineering, Oracle
Intel® Xeon® Processor E5-2600 v3 Product Family
12
The highest MySQL in-memory performance, scalability, and availability for
web, cloud and telecommunications services.
Intel® Xeon® processor E5-2600 v3 with enhanced microarchitecture,
additional enhanced cores and larger cache boosts the performance of
database applications such as MySQL Cluster*.
More queries and faster response times – a new performance record for
MySQL Cluster 7.4!
Over 200 Million QPS with MySQL Cluster
Core Business Applications
1 - Testing conducted on MySQL Cluster* 7.4 comparing Intel® Xeon® Processor E5-2697 v2 with Intel® Xeon® Processor E5-2697 v3. Testing done by Intel. For complete configuration details see slide 41.
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.
XPerformanceIncrease
MySQL Cluster* 7.4 increased performance (117M
to 156M reads/second) with the Intel® Xeon®
processor E5-2697 v31
Up to 33%
faster!
www.mysql.com/products/cluster/
INTEL CONFIDENTIAL – USE UNDER NDA ONLY *Other names and brands may be claimed as the property of others
N E W
16. 16
Infrastructure is aging and
holding companies back.
IT must enable business
innovation while also cutting
costs.
This creates a key challenge
for business competitiveness
and growth.
Economic
Headwinds Have
Put Pressure on
IT Budgets
17. 32% of Servers are
>4 Years Old1
32%
>4YEARSOLD
Make up only 4% of
Total Performance
Capabilities of Servers1 4%
>4YEARSOLD
Use 65% of Total
Energy Consumption1 65
>4YEARSOLD
%
26%
CAPEX
74%
OPEX
Business innovation throttled to 26%2
Time to revenue
Cost of lost time, effort, opportunity
Unpredictable business cycles
74% captive in operations and maintenance2
Rigid & aging infrastructure
Application & information complexity
Inflexible business processes
1*Source:Intelanalysis,2012
2*Source:Gartner,ITMetrics:AlignITInvestmentLevelsWithStrategyUsingRun,Grow,TransformandBeyond
(March2012)
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausethe
resultstovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.
Formoreinformationgotohttp://www.intel.com/performance
Aging Infrastructure is Inefficient and Costly
18. EXECUTE
• Configure new hardware and install
new operating system and
applications
• Test functionality and acceptance
with select users
• Establish backup and disaster
recovery functionality
• Release to users
• Plan for new upgrade cycle
An Approach to Infrastructure Modernization
PLAN
• Determine site requirements for
hosting hardware
• Assign applications and development
environments to hardware
• Define network architecture and
storage requirements
• Assess ability of new architecture to
meet compute requirements
• Develop project plan with timeline
and costs
EXPLORE
• Determine business requirements
• Identify utilization and growth in
servers, network, and storage
• Identify software (custom or COTS)
• Diagram infrastructure architecture
• Determine costs and cost trends
MODEL
• Compare business requirements to
vendor offerings
• Identify application interoperability
issues across the solution stack
• Determine where applications will
be hosted (virtual, physical, cloud)
• Define the hardware components
to retain, upgrade, and acquire
19. The Journey Starts with a Strong Foundation
• Take advantage of new capabilities in hardware and software to support business growth
• Minimize operating expenses, maximize efficiency
• Ensure security and compliance
19
20. Effective IT
Empowers
Business Success
Companies that have invested
in IT to achieve operational
excellence and innovation are
seeing tremendous benefits.
Companies that integrate IT
into the business to deliver
differentiated services and
value for their customers are
thriving.
20
21. Increase Operational
Efficiency
Minimize operating expenses
and maximize efficiency
Take advantage of new capabilities in
hardware and software
Ensure security and compliance
Ensure reliability to enhance service
quality (nonstop services, VoIP, XoIP, etc.)
The Path to
Transformational IT
Data Center Optimization
Open
Standards
Based
Modernize
Service Delivery
Deliver new services on demand
Use private, public, or hybrid
cloud
Develop APIs and service-
oriented architecture
Create New
Business Opportunities
Use big data to identify
opportunities and respond to
competitive threats
Optimize operations that improve
time to market and predictive
analytics
Enhance customer value with new
products and services
2121
22. 22
Solutions that Work Better Together
Oracle JAVA*
Standard Edition 8
Intel® Xeon®
Processor E5-2600
v3 Product Family
2014
2010
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausethe
resultstovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.
Formoreinformation,visitintel.com/performance
*1,2.3,4+Configurationdetailsandadditionalinformationonthefollowingpage.
2014
Baseline
• Oracle* JDK 7
• Intel® Xeon® Processor
X5690
Software Upgrade
+ Oracle* JDK 8
Optimized JVM
• Improved memory handling
• Up to 1.25x performance1
Better Together
Up to 8.7x performance
improvement4
• Support business
transaction peak times,
meet SLAs
• Increase datacenter
efficiency
• Headroom for future
growth
Workload: SPECjbb*2013-MultiJVM
Hardware Upgrade
+ Intel® Xeon® Processor E5-2699 v3
Increased parallelism - up to 18 cores
• Up to 3.5x performance2
• Up to 2.7x memory capacity3
*Other names and brands may be claimed as the property of others.
24. 24
Solutions that Work Better Together
Microsoft SQL*
2014
Intel® Xeon®
Processor E5-2600
v3 Product Family
2014
2010
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausethe
resultstovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.
Formoreinformation,visitintel.com/performance
*1,2.3+Configurationdetailsandadditionalinformationonthefollowingpage.
2014
Baseline
• Microsoft SQL Server*
2008 R2
• Intel® Xeon® Processor
X5680
Software Upgrade
+ Microsoft SQL Server* 2014
Performance and Scalability enhancements:
• In-Memory Online Transaction Processing
• Faster query response times
Better Together
Up to 2.8xperformance3
Delivers in-memory analytics
and business insight through a
modern, fully-integrated
enterprise-class data platform
Workload: brokerage firm OLTP
Hardware Upgrade
+ Intel® Xeon® Processor E5-2699 v3
Increased parallelism - up to 18 cores
• Up to 2.7x memory capacity1
The DDR4 Difference
• Up to 3x increased memory bandwidth2
*Other names and brands may be claimed as the property of others.
25. 25
Configuration Details for Microsoft SQL Server* and Intel® Xeon® Processor E5-2600 v3 Product Family
1* Up to 2.7X memory capacity based on a typical 2-socket server platform configuration: Intel® Xeon® processor X5600 series supports 18 DIMMS, max memory per DIMM of 32 GB RDIMM;
Intel® Xeon® processor E5-2600 v3 family supports 24 DIMMs, max memory per DIMM of 64GB RDIMM.
2* Up to 3x memory bandwidth based on STREAM(triad) benchmark comparing baseline Supermicro X8DTN+ platform with two Intel® Xeon® Processor X5680, 18x8GB DDR3-800 scoring
26.5 GB/sec to the new Intel® Server System R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3, 24x16GB DDR4-2133 @ 1600MHz DR-RDIMM scoring 85.2 GB/sec. Source: Intel
internal testing.
3* Up to 2.8x performance based on internal SQL Server* OLTP workload comparing baseline Supermicro X8DTN+ platform with two Intel® Xeon® Processor X5680 using SQL Server 2008 R2
scoring 1.256 Ktps (source: Intel technical report #1204) to the new Intel® Server System R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3 using SQL Server 2014 scoring 3.567 Ktps
(source: Intel technical report #3049).
Processor
Intel® Xeon® Processor X5690
(6C, 3.46 GHz, 130W)
Intel® Xeon® Processor E5-2699 v3
(18C, 2.3 GHz, 145W)
Sockets 2 2
Memory 144 GB (18 x 8 GB PC3-10600) 512GB (16 x 8GB DDR4-2133)
Disk Intel® SSD X25 Series Intel® SSD DC S3700 Series
OS Distribution Microsoft Windows Server* 2008 Microsoft Windows Server* 2012
Software Microsoft SQL Server* 2008 R2 Microsoft SQL* 2014
OLTP Throughput 1.256 K transactions per second 3.567 K transactions per second
Source Intel Technical Report #1204 Intel Technical Report #3049
*Other names and brands may be claimed as the property of others.
26. 26
Solutions that Work Better Together
Microsoft SQL*
2014
Intel® Xeon®
Processor E5-2600
v3 Product Family
2014
2012
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausethe
resultstovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.
Formoreinformation,visitintel.com/performance
*1,2+Configurationdetailsandadditionalinformationonthefollowingpage.
2014
Baseline
• Microsoft SQL 2012*
• Intel® Xeon® Processor
E5-2697 v2
Software Upgrade
+ Microsoft SQL Server 2014
Performance and Scalability
enhancements:
• In-Memory Online Transaction
Processing
• Faster query response times
Better Together
Up to 1.4xperformance3
Delivers in-memory analytics
and business insight through a
modern, fully-integrated
enterprise-class data platform
Workload: brokerage firm OLTP
Hardware Upgrade
+ Intel® Xeon® Processor E5-2699 v3
Increased parallelism - up to 18 cores
Up to 2x memory capacity1
*Other names and brands may be claimed as the property of others.
27. 27
Configuration Details for Microsoft SQL Server* and Intel® Xeon® Processor E5-2600 v3 Product Family
1* Up to 2X memory capacity based on a 2-socket platform: Intel® Xeon® processor E5-2600 series supports 24 DIMMS, max memory per DIMM of 32 GB RDIMM; Intel® Xeon® processor E5-
2600 v3 family supports 24 DIMMs, max memory per DIMM of 64 GB RDIMM.
2* Up to 1.4x performance based on internal SQL Server* OLTP workload comparing baseline Intel® Internal platform with two Intel® Xeon® Processor E5-2697 v2 using SQL Server 2012
scoring 2.472 Ktps (source: Intel technical report #3409) to the new Intel® Server System R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3 using SQL Server 2014 scoring 3.567
Ktps (source: Intel technical report #3049).
Processor
Intel® Xeon® Processor E5-2697 v2
(12C, 2.7 GHz, 130W)
Intel® Xeon® Processor E5-2699 v3
(18C, 2.3 GHz, 145W)
Sockets 2 2
Memory 256 GB (16x16 GB DDR3-1600) 512GB (16 x 32GB DDR4-2133)
Disk Intel® SSD DC S3700 Series Intel® SSD DC S3700 Series
OS Distribution Microsoft Windows Server* 2012 Microsoft Windows Server* 2012
Software Microsoft SQL* 2012 Microsoft SQL* 2014
OLTP Throughput 2.472 K transactions per second 3.567 K transactions per second
Source Intel Technical Report #3049 Intel Technical Report #3049
*Other names and brands may be claimed as the property of others.
28. 28
Get the Most Out of Modern Software & Hardware
Intel® Xeon® E5-2600 v3 product family and SQL Server* 2014
Microsoft Windows*
Server 2008 R2 and
SQL Server* 2008
Six-Core Intel® Xeon®
Processor X5690
Based Servers
100
&
Softwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonly
onIntelmicroprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasured
usingspecificcomputersystems,components,software,operationsandfunctions.Anychangeto
anyofthosefactorsmaycausetheresultstovary.Youshouldconsultotherinformationand
performanceteststoassistyouinfullyevaluatingyourcontemplatedpurchases,includingthe
performanceofthatproductwhencombinedwithotherproducts.
Formoreinformation,visitintel.com/performance
Configurationdetailsonthefollowingpage
Up to
increased throughput1
2.8x
Eighteen-Core
Intel® Xeon®
Processor
E5-2600 v3
Product Family-
based Servers
100
Business Value
Performance
Performance Benefits
Business Benefits
In-Memory Online Transaction Processing
leads to faster query response times.
Get to insights faster with a complete BI platform
that speeds up how you access, analyze, clean and
shape both internal and external data
*Other names and brands may be claimed as the property of others.
29. 29
Configuration Details for Microsoft SQL Server* and Intel® Xeon® Processor E5-2600 Family
1 Up to 2.8x performance based on internal SQL Server* OLTP workload comparing baseline Supermicro X8DTN+ platform with two Intel® Xeon® Processor X5680 using SQL Server 2008 R2
scoring 1.256 Ktps (source: Intel technical report #1204) to the new Intel® Server System R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3 using SQL Server 2014 scoring 3.567 Ktps
(source: Intel technical report #3049).
Processor
Intel® Xeon® Processor X5690
(6C, 3.46 GHz, 130W)
Intel® Xeon® Processor E5-2699 v3
(18C, 2.3 GHz, 145W)
Sockets 2 2
Memory 144 GB (18 x 8 GB PC3-10600) 512GB (16 x 8GB DDR4-2133)
Disk Intel® SSD X25 Series Intel® SSD DC S3700 Series
OS Distribution Microsoft Windows Server* 2008 Microsoft Windows Server* 2012
Software Microsoft SQL Server* 2008 R2 Microsoft SQL* 2014
OLTP Throughput 1.256 K transactions per second 3.567 K transactions per second
Source Intel Technical Report #1204 Intel Technical Report #3049
*Other names and brands may be claimed as the property of others.
30. 30
Solutions that Work Better Together
Microsoft Windows
Server* 2012
Intel® Xeon®
Processor E5-2600
v3 Product Family
HPC
2014
2010
2014
Baseline
• Windows Server* 2003
• Intel® Xeon® Processor
X5690
Software Upgrade
+ Microsoft Windows Server* 2012 R2
• Support up to 5x logical processors to
massively increase parallelizability
• Enhanced power management
interface offers improved control over
power consumption
Better Together
Up to 3.4xhigher
performance3
• Increased energy efficiency
across all load levels
• Intel and Microsoft* deliver
industry leading
performance across a range
of real-world high
performance computing
applications
Benchmark: SPECfp*_rate_base2006
Hardware Upgrade
+ Intel® Xeon® Processor E5-2699 v3
• Increased parallelism with up to 18 cores
• Intel® AVX2® delivers up to 4x DP
FLOPS/core and supports 2x wider vector
integer instructions1
• The DDR4 difference: Up to 3x more
memory bandwidth2
*Other names and brands may be claimed as the property of others.
SPECfp*rateisatrademarkoftheStandardPerformanceEvaluationCorporation(SPEC).
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausethe
resultstovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.
Formoreinformation,visitintel.com/performance
*1,2,3+Configurationdetailsandadditionalinformationonthefollowingpage.
31. 31
Configuration Details for Microsoft Windows Sever* 2012 and Intel® Xeon® Processor
E5-2600 v3 Product Family
*Other names and brands may be claimed as the property of others.
1* Intel® Xeon® processor 5600 series, Intel AVX enables up to twice the work to be accomplished per clock cycle during floating point and vector operations
2* Up to 3x memory bandwidth based on STREAM(triad) benchmark comparing baseline Supermicro X8DTN+ platform with two Intel® Xeon® Processor X5680, 18x8GB DDR3-800 scoring 26.5 GB/sec
to the new Intel® Server System R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3, 24x16GB DDR4-2133 @ 1600MHz DR-RDIMM scoring 85.2 GB/sec. Source: Intel internal testing.
3* Up to 3.4x performance improvement based on SPECfp*_rate_base2006 using different Intel® Compilers comparing baseline Supermicro X8DTN+ with two Intel® Xeon® Processor X5690
(http://www.spec.org) scoring 226 to the Intel® Server Platform with two Intel® Xeon® Processor E5-2699 v3 scoring 789. Source: Intel internal estimates.
4* Idle power represents the average power of the server while not under any workload.
Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual
performance. Physical to physical consolidation scenario. 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 SSE3 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
Processor
Intel® Xeon® Processor
X5690
(6C, 3.46 GHz, 130W)
Intel® Xeon® Processor
X5690
(6C, 3.46 GHz, 130W)
Intel® Xeon® Processor
E5-2699 v3
(18C, 2.3 GHz, 145W)
Intel® Xeon® Processor
E5-2699 v3
(18C, 2.3 GHz, 145W)
Sockets 2 2 2 2
Memory
48 GB
(16x4 GB DDR3-1600)
48 GB
(16x4 GB DDR3-1600)
128GB
(16 x 8GB DDR4-2133)
128GB
(16 x 8GB DDR4-2133)
OS Distribution Windows Server* 2003 Windows Server* 2012 R2 Windows Server* 2003 Windows Server* 2012 R2
Compiler Intel® Compiler 12.1.0.255 Intel® Compiler 14.0 Intel® Compiler 12.1.0.255 Intel® Compiler 14.0
Idle Power (watts) 165 (Intel Est.4) 119(Intel Est.4)
119 (Intel Est.4)
86 (Intel Est.4)
Performance
(SPECfp_rate_base2006)
226 (Intel Est.3) 270 (Intel Est.3) 660 (Intel Est.3) 789 (Intel Est.3)
32. 32
Solutions that Work Better Together
Microsoft Windows
Server* 2012
Intel® Xeon®
Processor E5-2600
v3 Product Family
Enterprise
2014
2010
2014
Baseline
• Windows Server* 2003
• Intel® Xeon® Processor
X5690
Software Upgrade
+ Microsoft Windows Server* 2012 R2
• Support up to 5x logical processors to
massively increase parallelizability
• Enhanced power management
interface offers improved control over
power consumption
Better Together
Up to 3.1xhigher
performance3
• Increased energy efficiency
across all load levels
• Intel and Microsoft* deliver
industry leading
performance across a range
of real-world high
performance computing
applications
Benchmark: SPECint*_rate_base2006
Hardware Upgrade
+ Intel® Xeon® Processor E5-2699 v3
• Increased parallelism with up to 18 cores
• Intel® AVX2® delivers up to 4x DP
FLOPS/core and supports 2x wider vector
integer instructions1
• The DDR4 difference: Up to 3x more
memory bandwidth2
*Other names and brands may be claimed as the property of others.
SPECint*rateisatrademarkoftheStandardPerformanceEvaluationCorporation(SPEC).
SoftwareandworkloadsusedinperformancetestsmayhavebeenoptimizedforperformanceonlyonIntel
microprocessors.Performancetests,suchasSYSmarkandMobileMark,aremeasuredusingspecificcomputer
systems,components,software,operationsandfunctions.Anychangetoanyofthosefactorsmaycausethe
resultstovary.Youshouldconsultotherinformationandperformanceteststoassistyouinfullyevaluatingyour
contemplatedpurchases,includingtheperformanceofthatproductwhencombinedwithotherproducts.
Formoreinformation,visitintel.com/performance
*1,2,3+Configurationdetailsandadditionalinformationonthefollowingpage.
33. 33
Configuration Details for Microsoft Windows Sever* 2012 and Intel® Xeon® Processor E5-
2600 v3 Product Family
*Other names and brands may be claimed as the property of others.
1*The Intel® Xeon® processor E5 product family supports Intel® Advanced Vector Extensions (Intel® AVX), which increases maximum vector size from 128 to 256 bits. Compared to the Intel® Xeon®
processor 5600 series, Intel AVX enables up to twice the work to be accomplished per clock cycle during floating point and vector operations
2* Up to 3x memory bandwidth based on STREAM(triad) benchmark comparing baseline Supermicro X8DTN+ platform with two Intel® Xeon® Processor X5680, 18x8GB DDR3-800 scoring 26.5 GB/sec
to the new Intel® Server System R2208WTTYS with two Intel® Xeon® Processor E5-2699 v3, 24x16GB DDR4-2133 @ 1600MHz DR-RDIMM scoring 85.2 GB/sec. Source: Intel internal testing.
3* Up to 3.1x performance improvement based on SPECint*_rate workload comparing baseline Supermicro X8DTN+ with two Intel® Xeon® Processor X5690 (source link) using Intel® Compiler
12.1.0.255 scoring 369 to the Intel® Wildcat Pass server with two Intel® Xeon® Processor E5-2699 v3 using Intel® Compiler 14.0 scoring 1170.
4* Idle power represents the average power of the server while not under any workload.
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
SSE3 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 .
Processor
Intel® Xeon® Processor
X5690
(6C, 3.46 GHz, 130W)
Intel® Xeon® Processor
X5690
(6C, 3.46 GHz, 130W)
Intel® Xeon® Processor
E5-2699 v3
(18C, 2.3 GHz, 145W)
Intel® Xeon® Processor
E5-2699 v3
(18C, 2.3 GHz, 145W)
Sockets 2 2 2 2
Memory
48 GB
(16x4 GB DDR3-1600)
48 GB
(16x4 GB DDR3-1600)
128GB
(16 x 8GB DDR4-2133)
128GB
(16 x 8GB DDR4-2133)
OS Distribution Windows Server 2003 Windows Server 2012 R2 Windows Server 2003 Windows Server 2012 R2
Compiler Intel® Compiler 12.1.0.255 Intel® Compiler 14.0 Intel® Compiler 12.1.0.255 Intel® Compiler 14.0
Idle Power (watts) 165 (Intel Est.4) 111.14 (Intel Est.4)
119 (Intel Est.4)
86 (Intel Est.4)
Performance
(SPECint_rate_base2006)
369 (Intel Est.3) 406 (Intel Est.3) 1063 (Intel Est.3) 1170 (Intel Est.3)
34. 34
IBM DB2* Solution Brief
Enterprise Database
Oracle (JAVA JDK8)* Solution Brief
Programming and Runtime Environment
See more Core Business Applications and Intel® Xeon® processor E5-2600 v3
stories at http://transformingbusiness.intel.com/software-solutions
35. 35
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. For more information go to
http://www.intel.com/performance Intel does not control or audit third-party benchmark data or the web sites referenced in this document. You should visit the referenced web site and confirm whether referenced data are accurate.
*Other names and brands may be claimed as the property of others.
1. Source as of September 8, 2014. New configuration: Hewlett-Packard Company HP ProLiant ML350 Gen9 platform with two Intel® Xeon® Processor E5-2699 v3, Oracle Java Standard Edition 8 update 11, 190,674 SPECjbb2013-MultiJVM max-jOPS, 47,139 SPECjbb2013-MultiJVM critical-jOPS.
Source. Baseline: Cisco Systems Cisco UCS C240 M3 platform with two Intel® Xeon® Processor E5-2697 v2, Oracle Java Standard Edition 7 update 45, 63,079 SPECjbb2013-MultiJVM max-jOPS , 23,797 SPECjbb2013-MultiJVM critical-jOPS. Source.
Accelerating Business Transformation
Up to 3.0X performance improvement1 for
efficient, high performance solutions
Common architecture for
compute, storage and networks
Monitoring and control for
improved system visibility
The Foundation for the
Software-Defined Infrastructure
Intel® Xeon® Processor. Intel® Solid-State Drives. Intel® Ethernet Solutions. Intel®
Software Development Tools
36. Product and Support Pages
Intel® Xeon® processor E5 v3 product family:
• Product page: http://www.intel.com/content/www/us/en/processors/xeon/xeon-
processor-e5-family.html
• Support page: http://www.intel.com/p/en_US/support/highlights/processors/xp-e5-2600
Intel® Solid-State Drive Data Center P3700:
• Product Page: http://www.intel.com/content/www/us/en/solid-state-drives/intel-ssd-dc-
family-for-pcie.html
• Support Page: http://www.intel.com/p/en_US/support/highlights/ssdc/ssd-dc-p3700
Intel® Ethernet Solutions:
• Product Page: http://www.intel.com/content/www/us/en/network-adapters/converged-
network-adapters.html
• Support Page: http://www.intel.com/p/en_US/support/highlights/network/xl710
Intel® Server Software for Code Vectorization, Parallelism and Optimization:
• Product Page: https://software.intel.com/en-us/intel-mkl
• Product Page: https://software.intel.com/en-us/intel-parallel-studio-xe
39. 39
Intel® Advanced Vector Extensions 2 (Intel® AVX2): Intel® Advanced
Vector Extensions 2 (Intel® AVX2) provides higher throughput to
certain processor operations. Due to varying processor power
characteristics, utilizing AVX instructions may cause a) some parts
to operate at less than the rated frequency and b) some parts with
Intel® Turbo Boost Technology 2.0 to not achieve any or maximum
turbo frequencies. Performance varies depending on hardware,
software, and system configuration and you can learn more at
http://www.intel.com/go/turbo.
Legal Disclaimers: Performance
Intel® Hyper-Threading Technology (Intel® HT Technology): Available
on select Intel® processors. Requires an Intel® HT Technology-
enabled system. Your performance varies depending on the specific
hardware and software you use. Learn more by visiting
http://www.intel.com/info/hyperthreading.
41. 41
Configuration Details:
IBM WEBSPHERE APPLICATION SERVER*: The workload uses standards-based HTTP and Web services protocols such as SOAP and WSDL. DayTrader*
provides the EJB server implementations of the emulated "Trade" brokerage services. The setup has 4 nodes: 2 drivers, 1 application server, and 1 database
machine (uses DB2). 10G connections are used between the app server and the drivers (3 NICs).
Baseline configuration: Websphere 8.5.5/Daytrader* 3 on SUSE Enterprise Linux* 11 SP1, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 12 cores, 64GB
DDR3/1600, regular DIMM, Intel® SSD DC P3700, Intel® 82599ES 10 Gigabit Ethernet Controller, Source is Intel internal as of July 18, 2014.
New configuration: Websphere 8.5.5/Daytrader* 3 on SUSE Enterprise Linux* 11 SP1, 2 Intel® Xeon® processor E5-2697 v3, 2.6GHz, 14 cores, 64GB
DDR4/2133, regular DIMM, Intel® SSD DC P3700, Intel® 82599ES 10 Gigabit Ethernet Controller, Source is Intel internal as of July 18, 2014.
IBM INFOSPHERE DATASTAGE* v9.1.2*: Typical Data Warehousing Integration Scenario; based on Brokerage Firm Business Scenario that performs extract,
transform and load operations (ETL). CPU intensive ETL test (DimTrade) used. Data volume size is 13GB.
Baseline configuration: DataStage 9.1.2 on Red Hat* Enterprise Linux 6.5, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 12 cores, 64GB DDR3/1600, 1TB
7200 RPM, 1 GbE network, Source is Intel internal as of July 18, 2014.
New configuration: DataStage 9.1.2 on Red Hat* Enterprise Linux 6.5, 2 Intel® Xeon® processor E5-2697 v3, 2.6GHz, 14 cores, 64GB DDR4/2133, 1TB 7200
RPM, 1 GbE network, Source is Intel internal as of July 18, 2014.
NOMURA RESEARCH INSTITUTE BOS*: Real-time Stock calculation (Normalized TPC-H benchmark schema with Bill of Materials). Deploy a Java Virtual
Machine and two PostgreSQL database instances for each NUMA node. Data Manipulation on top of extremely Normalized database schema (INSERT ONLY
design constraint without UPDATE and DELETE operations).
Baseline: RHEL 7, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 12 cores, 64GB/1867, DDR3, 2 SAS disks, 10GB network speed
New: RHEL 7, 2 Intel® Xeon® processor E5-2697 v3, 2.6GHz, 18 cores, 64GB/2133, DDR4, 2 SAS disks, 10GB network speed. Testing done by Nomura
Research Institute 2/10/2015
MARIADB CORPORATION*: Simulates typical R/O OLTP requests, 1M table size and 10M RW. Multi-table scenario.
Baseline configuration: MariaDB* 10.1 on Ubuntu* 14.04 x64, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 12 cores, 64GB/1867, DDR3, 2 SATA/7200RPM,
1Gb network, Source is Intel internal as of September 7, 2014.
New configuration: MariaDB* 10.1 on Ubuntu* 14.04 x64, 2 Intel® Xeon® processor E5-2697 v3, 2.6GHz, 14 cores, 64GB/2133, DDR4, 2 SATA/7200RPM, 1Gb
network, Source is Intel internal as of September 7, 2014.
ORACLE MYSQL CLUSTER* 7.4: flexAsynch benchmark
Testing conducted on MySQL Cluster* 7.4 comparing Intel® Xeon® Processor E5-2697 v2 with Intel® Xeon® Processor E5-2697 v3. Testing done by Intel.
42. 42
YONYOU*: UFIDA U9* workload uses stress tool to simulate typical business operations in U9 and the pressure focuses on application server side. CPU
utilization is around 85% and performance metric is TPS (Transactions per second).
Baseline configuration: UFIDA U9* on Windows Server* 2012 R2, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 24 cores, 32GB DDR3/1600, 1 SATA Disk, 1
GbE network, Source is Intel internal as of June 24, 2014.
New configuration: UFIDA U9* on Windows Server* 2012 R2, 2 Intel® Xeon® processor E5-2697 v3, 2.6GHz, 28 cores, 32GB DDR3/2133, 1 SATA Disk, 1 GbE
network, Source is Intel internal as of June 24, 2014.
IBM DB2*: Proof of Performance and Scalability (POPS) is a data warehousing workload based on a true retail environment. The workload operates on a star
schema database. A 2TB database (based on raw data) was used for this test, and the results are an aggregation of 36 queries (10 fact-to-fact joins and 13
snowflake queries). All queries were run in single user mode.
Baseline configuration: IBM DB2* 10.5 FP4 on SUSE Enterprise Linux* 11.3 SP1, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 12 cores, 256GB DDR3/1333,
132 SAS disks, IBM XIV storage, 1 GbE network, Source is Intel internal as of August 12, 2014.
New configuration: IBM DB2* 10.5 FP4 on SUSE Enterprise Linux* 11.3 SP1, 2 Intel® Xeon® processor E5-2699 v3, 2.3GHz, 18 cores, 64GB DDR4/2133, 132
SAS disks, IBM XIV storage, 1 GbE network, Source is Intel internal as of August 12, 2014.
MySQL*: Simulates typical R/O OLTP requests, 1M table size and 10M RW.
Baseline configuration: MySQL* 5.7.4 on Ubuntu* 14.04 x64, 2 Intel® Xeon® processor E5-2697 v2, 2.7GHz, 12 cores, 64GB/1866, DDR3, 2 SATA/7200RPM,
1Gb network, Source is Intel internal as of August 14, 2014.
New configuration: MySQL* 5.7.4 on Ubuntu* 14.04 x64, 2 Intel® Xeon® processor E5-2697 v3, 2.6GHz, 14 cores, 64GB/2133 DDR4, 2 SATA/7200RPM, 1Gb
network, Source is Intel internal as of August 14, 2014.
Key Message:
Organizations today are challenged to find the right balance between cutting costs and deploying modern IT infrastructure to support ever-evolving market demands.
Key Message:
Net-Net: 32% of the world’s servers only contribute 4% of the total performance but use 65% of the total power. These servers are greater than 4 years old
Key Message:
Modernizing the IT infrastructure is a process. Here is a diagram showing an application modernization methodology, and the key phases and considerations. This is an ongoing cycle that should be completed every 2 to 4 years.
Key Message:
Start by updating your IT infrastructure to minimize operating expenses and maximize efficiency, taking advantage of new capabilities in hardware and software.
Additional Detail:
Extraordinary demands require extraordinary innovation
Intel is actively working with the industry to come up with creative and innovative solutions to meet this insatiable demand for compute in high performance computing
Intel is investing in three key areas… New CPU architectures, New Software & Tools, and Fabrics… …for a balanced platform approach… designed to keep up with the aggressive pursuit of computational capability
Key Message:
Organizations that utilize IT as a strategic enabler for operational excellence and differentiated services have a distinct competitive advantage.
Key Message:
There is a clear path to IT leadership: It starts with getting your IT environment in order and reducing costs while maximizing performance, security, and compliance.
Delivering a cloud infrastructure compounds costs savings, while enabling a flexible, service-based IT model.
From there, you can create new business opportunities and embrace new workloads like big data.
Additional Detail:
What is your balance in these three buckets?
Do you think of your IT in these 3 buckets?
How do you free up $$ from operating expenses to focus on modernized service delivery and new business opportunities?