Intel Dual-core Intel Xeon 3085 BX805573085 User Manual
Product codes
BX805573085
7
• Benchmark Description for SPECcpu*006 suite (SPECint*_rate_base006 & SPECfp*_rate_base006): SPEC CPU2006 is the industry-adopted, CPU-intensive benchmark which
stresses the system processor(s), memory subsystem, and compiler. Derived from 29 real user applications, CPU2006 provides a comparison across the widest practical range of hardware
reporting a geometric mean ratio score on a baseline compiled binary.
reporting a geometric mean ratio score on a baseline compiled binary.
• Best 1-socket server SPECint_rate_base006* (score) benchmark results as of 1 Oct 007:
– Intel® Pentium® D Processor platform (score 20.6): Fujitsu-Siemens PRIMERGY* RX100S4 using Intel Pentium D processor 945 (3.40 GHz, 2x2 MB L2 cache, 800 MHz bus), 8 GB
(4x 2 GB DDR2 PC4200E, 2-rank, CL4, ECC), 64-bit SuSE LINUX* Enterprise Server 10 2.6.16.21-08smp, Intel® C++ Compiler for IA32/EM64T version 9.1.
Source: http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070619-01287.html.
Source: http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070619-01287.html.
– Dual-Core Intel® Xeon® Processor 3070 platform (score 29.3): Fujitsu-Siemens PRIMERGY* RX100S4 using Intel Xeon processor 3070 (2.66 GHz, 4 MB L2 cache, 1066 MHz bus),
8 GB (4x 2 GB DDR2 PC4200E, 2-rank, CL4, ECC), 64-bit SuSE LINUX* Enterprise Server 10 2.6.16.21-08smp, Intel® C++ Compiler for IA32/EM64T version 9.1.
Source: http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070514-01066.html.
Source: http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070514-01066.html.
– Dual-Core Intel® Xeon® Processor 3085 platform (score 35.3): Intel® SR1530HSH using Intel Xeon processor 3085 (3.00 GHz, 4 MB L2 cache, 1333 MHz bus), 8 GB (4x 2GB DDR2-800),
SUSE Linux10* SP1 RC5, Intel® C++ Complier Compiler version 10.1. Source: Intel internal measurements TR#824.
– Quad-Core Intel® Xeon® Processor X3230 platform (score 48.5): Intel® SR1530HSH using Intel Xeon processor X3230 (2.66 GHz, 8 MB L2 cache, 1066 MHz bus), 8 GB
(4x 2GB DDR2-800), SUSE Linux10* SP1 RC5, Intel® C++ Complier Compiler version 10.1. Source: Intel internal measurements TR#824.
• Best 1-socket server SPECfp_rate_base006* (score) benchmark results as of 1 Oct 007:
– Intel® Pentium® D Processor platform (score 19.8): Fujitsu-Siemens PRIMERGY* RX100S4 using Intel Pentium D processor 945 (3.40 GHz, 2x2 MB L2 cache, 800 MHz bus), 8 GB
(4x 2 GB DDR2 PC4200E, 2-rank, CL4, ECC), 64-bit SuSE LINUX* Enterprise Server 10 2.6.16.21-08smp, Intel® C++ Compiler for IA32/EM64T version 9.1.
Source: http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070619-01288.html.
Source: http://www.spec.org/cpu2006/results/res2007q3/cpu2006-20070619-01288.html.
– Dual-Core Intel® Xeon® Processor 3070 platform (score 25.3): Acer Altos* G330 using Intel Xeon processor 3070 (2.66 GHz, 4 MB L2 cache, 1066 MHz bus), 4 GB
(4x 1 GB DDR2-667, CL5, ECC), 64-bit SuSE LINUX* Enterprise Server 10 2.6.16.21-08smp, Intel® C++ Compiler for IA32/EM64T version 9.1.
Source: http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070429-00964.html.
Source: http://www.spec.org/cpu2006/results/res2007q2/cpu2006-20070429-00964.html.
– Dual-Core Intel® Xeon® Processor 3085 platform (score 29.7): Intel® SR1530HSH using Intel Xeon processor 3085 (3.00 GHz, 4 MB L2 cache, 1333 MHz bus), 8 GB (4x 2GB DDR2-800),
SUSE Linux10* SP1 RC5, Intel® C++ Complier and Fortran Compiler version 10.1. Source: Intel internal measurements TR#824.
– Quad-Core Intel® Xeon® Processor X3230 platform (score 35.2): Intel® SR1530HSH using Intel Xeon processor X3230 (2.66 GHz, 8 MB L2 cache, 1066 MHz bus), 8 GB
(4x 2GB DDR2-800), SUSE Linux10* SP1 RC5, Intel® C++ Complier and Fortran Compiler 10.1. Source: Intel internal measurements TR#824.
• Benchmark Description for SPECjbb*00: SPEC Java Business Benchmark 2005 (jbb2005). Written in Java, this multi-threaded benchmark emulates an order processing environment
in a company with multiple warehouses serving multiple customers. Measures average transaction throughput of a heavily loaded server. Performance reported in Business Operations per
Second (BOPS).
Second (BOPS).
• Best 1-socket server SPECjbb00* (bops) benchmark results as of 1 Oct 007:
– Intel® Pentium® D Processor platform (score 34,111): Fujitsu-Siemens PRIMERGY* TX150S4 using Intel® Pentium® D Processor 950 (3.40 GHz, 800 MHz bus, 2x 2 MB L2 cache),
8192 MB RAM, Microsoft Windows* Server 2003 Enterprise x64-Edition SP1 (64-bit), BEA JRockit* P26.0.0-35-59328-1.5.0_06-20060321-1333-win-x86_64.
Source: http://www.spec.org/osg/jbb2005/results/res2006q2/jbb2005-20060425-00108.html.
Source: http://www.spec.org/osg/jbb2005/results/res2006q2/jbb2005-20060425-00108.html.
– Intel® Xeon® Processor platform (3070 score 65,656/avg sys watts 132; 3085 score 73,275/avg sys watts 131; X3230 score 97,434/avg sys watts 196): Intel “Garlow” 3210 Chipset-
based server platform BIOS 8/31/2007 using Intel® Xeon® Processor 3070 (2.66 GHz, 4 MB cache, 1066 FSB), 3085 (3.00 GHz, 4 MB cache, 1333 FSB), and X3230 (2.66 GHz, 8 MB
cache, 1066 FSB), 4x 2 GB DDR2-800 ECC CL5 DIMMs, 120 GB SATA HDD, Microsoft Windows* Server 2003 Enterprise x64-Edition SP1 (64-bit), SPECjbb* V1.07, BEA JRockit* P27.4.0.
Source: Intel internal measurements TR#825.
cache, 1066 FSB), 4x 2 GB DDR2-800 ECC CL5 DIMMs, 120 GB SATA HDD, Microsoft Windows* Server 2003 Enterprise x64-Edition SP1 (64-bit), SPECjbb* V1.07, BEA JRockit* P27.4.0.
Source: Intel internal measurements TR#825.
– AMD Opteron* Processor platform (score 58,189/avg sys watts 162): Hewlett-Packard ML115G1* using AMD Opteron* Processor 1220SE (2.80 GHz), 4x 2 GB DDR2-800 ECC CL5
DIMMs, 120 GB SATA HDD, Microsoft Windows* Server 2003 Enterprise x64-Edition SP1 (64-bit), SPECjbb* V1.07, BEA JRockit* P27.4.0. Source: Intel internal measurements TR#825.
• Benchmark Description for Fluent*: Fluent is a commercial engineering application used to model computational fluid dynamics. The benchmark consists of 9 standard workloads orga-
nized into small, medium and large models. These comparisons use all but the largest of the models which does not fit into the 8GB of memory available on the platforms. The Rating, the
default Fluent metric, was used in calculating the ratio of the platforms by taking a geometric mean of the 8 workload ratings measured.
default Fluent metric, was used in calculating the ratio of the platforms by taking a geometric mean of the 8 workload ratings measured.
• Best 1-socket Fluent* version 6.3.33 (8 of the 9 standard, jobs/day) benchmark results as of 1 Oct 007:
– Intel® Xeon® Processor platform (3085 score 1611.1/avg sys watts 131; X3230 score 2247.7/avg sys watts 192): Intel “Garlow” 3210 Chipset-based server platform BIOS 8/31/2007
using Intel® Xeon® Processor 3085 (3.00 GHz, 4 MB cache, 1333 FSB) and X3230 (2.66 GHz, 8 MB cache, 1066 FSB), 4x 2 GB DDR2-667 ECC CL5 DIMMs, 120 GB SATA HDD, Red Hat
Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
– AMD Opteron* Processor platform (score 1287.2/avg sys watts 164): Hewlett-Packard ML115G1* using AMD Opteron* Processor 1220SE (2.80 GHz), 4x 2 GB DDR2-667 ECC CL5
DIMMs, 120 GB SATA HDD, Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
• Benchmark Description for LS-Dyna*: LS-DYNA is a commercial engineering application used in finite element analysis such as a car collision. The workload used in these comparisons is
called 3 Vehicle Collision and is publicly available from www.topcrunch.org/. The metric for the benchmark is elapsed time in seconds.
• Best 1-socket LS-Dyna* version mpp971.7600..398 (3 cars@30ms, jobs/day) benchmark results as of 1 Oct 007:
– Intel® Xeon® Processor platform (3085 score 6.63/avg sys watts 136; X3230 score 8.96/avg sys watts 194): Intel “Garlow” 3210 Chipset-based server platform BIOS 8/31/2007
using Intel® Xeon® Processor 3085 (3.00 GHz, 4 MB cache, 1333 FSB) and X3230 (2.66 GHz, 8 MB cache, 1066 FSB), 4x 2 GB DDR2-667 ECC CL5 DIMMs, 120 GB SATA HDD,
Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
– AMD Opteron* Processor platform (score 5.67/avg sys watts 167): Hewlett-Packard ML115G1* using AMD Opteron* Processor 1220SE (2.80 GHz), 4x 2 GB DDR2-667 ECC CL5
DIMMs, 120 GB SATA HDD, Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
• Benchmark Description for Amber*: a package of molecular simulation programs. The workload measures the number of problems solved per day (PS) using eight standard molecular
dynamic simulations. See http://amber.ch.ic.ac.uk/amber8.bench1.html for more information.
• Best 1-socket Amber* version 9 (8 standard, PS/day) benchmark results as of 1 Oct 007:
– Intel® Xeon® Processor platform (3085 score 305.5/avg sys watts 134; X3230 score 491.9/avg sys watts 204): Intel “Garlow” 3210 Chipset-based server platform BIOS
8/31/2007 using Intel® Xeon® Processor 3085 (3.00 GHz, 4 MB cache, 1333 FSB) and X3230 (2.66 GHz, 8 MB cache, 1066 FSB), 4x 2 GB DDR2-667 ECC CL5 DIMMs, 120 GB SATA
HDD, Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
HDD, Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
– AMD Opteron* Processor platform (score 212.8/avg sys watts 164): Hewlett-Packard ML115G1* using AMD Opteron* Processor 1220SE (2.80 GHz), 4x 2 GB DDR2-667 ECC CL5 DIMMs,
120 GB SATA HDD, Red Hat Enterprise LINUX* 64-bit. Source: Intel internal measurements TR#823.
SPECint*2006 and SPECfp*2006 benchmark tests reflect the performance of the microprocessor, memory architecture and compiler of a computer system on compute-intensive, 32-bit or
64-bit applications. SPEC benchmark tests results for Intel microprocessors are determined using particular, well-configured systems. These results may or may not reflect the relative per-
formance of Intel microprocessor in systems with different hardware or software designs or configurations (including compilers). Buyers should consult other sources of information, includ-
ing system benchmarks, to evaluate the performance of systems they are considering purchasing. SPECcpu*2006 suite including SPECint2006 and SPECfp2006 are registered trademarks
of the Standard Performance Evaluation Corporation (SPEC).
64-bit applications. SPEC benchmark tests results for Intel microprocessors are determined using particular, well-configured systems. These results may or may not reflect the relative per-
formance of Intel microprocessor in systems with different hardware or software designs or configurations (including compilers). Buyers should consult other sources of information, includ-
ing system benchmarks, to evaluate the performance of systems they are considering purchasing. SPECcpu*2006 suite including SPECint2006 and SPECfp2006 are registered trademarks
of the Standard Performance Evaluation Corporation (SPEC).
Server and HPC Performance