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.
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.
Intel does not control or audit the design or implementation of third party benchmarks or websites referenced in this document. Intel encourages all of its customers to visit the referenced websites or others where similar performance benchmarks are reported and confirm whether the referenced benchmarks are accurate and reflect performance of systems available for purchase.
Platform configurations: Discrete GFX: 1U server with two eight-core Intel® Xeon® processor E5-2650, 2.0 GHz, 20MB cache, 8GT/s QPI, 8x1GB DDR3-1333 memory, 1 hard drive, 1 power supply, 2 NVIDIA GTX680* graphics cards; Intel Xeon processor E3 v3: 3U microserver with 12 quad-core Intel Xeon processor E3-1285L v3, 3.1 GHz, 8MB cache, 4x1GB DDR3-1333, 1 hard drive, 1 power supply. Rack is 15kW rack tower. Power per server: 692W (Discrete GFX), 836W (E3v3). Servers per rack: 20 (Discrete GFX), 13 (E3v3). Concurrent transcode per GPU: 8 (Discrete GFX), 10 (E3v3). GTX680 transcode performance came from NVIDIA white paper at http://international.download.nvidia.com/webassets/en_US/pdf/GeForce-GTX-680-Whitepaper-FINAL.pdf. Intel E3 transcode performance came from Intel whitepaper: Intel Quick Sync Video Technology on Intel® Iris™/Iris Pro Graphics (5100+ Series and Intel HD Graphics (4200+ Series)—Flexible Transcode Performance and Quality Explained.
Baseline configuration: Fujitsu PRIMERGY* TX 120 S3 with one Intel® Xeon® processor E3-1260L (quad-core, 2.4 GHz, 8 MB L3 cache), Intel® Turbo Boost Technology disabled, Intel® Hyper-Threading Technology (Intel® HT Technology) enabled, 8 GB (2x 4-GB DDR3-1333 ECC DIMM), 64-GB SATA SSD, Microsoft Windows* 2008 R2 SP1, Java SE Runtime Environment (build 1.6.0_25), Java HotSpot 64-Bit Server VM (version 1.6.0_25). Source: http://spec.org/power_ssj2008/results/res2011q2/power_ssj2008-20110531-00379.html SPECpower_ssj2008* ssj_ops/Watt: 4697
Previous configuration: Fujitsu PRIMERGY TX120 S3p with one Intel Xeon processor E3-1265L v2 (8-MB Cache, 2.5 GHz), Intel HT Technology enabled, 8 GB (2x 4-GB DDR3-1600 ECC DIMM), 6 Gb/s SATA SSD, Microsoft Windows 2008 R2 SP1, Oracle Corporation Oracle Java HotSpot* 64-Bit Server VM on Windows, version 1.6.0_31. Source: http://www.spec.org/power_ssj2008/results/res2012q2/power_ssj2008-20120511-00463.html. SPECpower_ssj2008* ssj_ops/Watt: 6109
New configuration: Fujitsu PRIMERGY TX140 S2 with one Intel Xeon processor E3-1275L v3 (8-MB Cache, 2.7 GHz), Intel HT Technology enabled, 8 GB (2x 4-GB DDR3-1600 ECC DIMM), 250 GB HDD 2.5-inch SATA, Microsoft Windows 2008 R2 SP1, IBM J9 VM (build 2.6, JRE 1.7.0 Windows Server 2008 R2 amd64-64 20120322_106209 (JIT enabled, AOT enabled). Source: http://www.spec.org/power_ssj2008/results/res2013q4/power_ssj2008-20130906-00627.html SPECpower_ssj2008* ssj_ops/Watt: 6797