WHITE PAPER FUJITSU PRIMERGY SERVERS PERFORMANCE REPORT PRIMERGY RX100 S7

Transcription

1 WHITE PAPER PERFORMANCE REPORT PRIMERGY RX100 S7 WHITE PAPER FUJITSU PRIMERGY SERVERS PERFORMANCE REPORT PRIMERGY RX100 S7 This document contains a summary of the benchmarks executed for the PRIMERGY RX100 S7. The PRIMERGY RX100 S7 performance data are compared with the data of other PRIMERGY models and discussed. In addition to the benchmark results, an explanation has been included for each benchmark and for the benchmark environment. Version Contents Document history... 2 Technical data... 3 SPECcpu SPECjbb SPECpower_ssj OLTP STREAM LINPACK Literature Contact Fujitsu Technology Solutions 2011 Page 1 (24)

2 Document history Version 1.0 First report version including the benchmark chapters SPECcpu2006 Measurements with Pentium G620, Core i and i3-2120, Xeon E3-1220L, E3-1260L, E3-1220, E3-1230, E3-1240, E and E SPECjbb2005 Measurement with Xeon E SPECpower_ssj2008 Measurement with Xeon E3-1260L and 1 x SATA 2.5 SSD OLTP-2 Results for Pentium G620 as well as Core i3 and Xeon E3 processor series STREAM Measurements with Pentium G620, Core i3-2100, i as well as Xeon E3-1220L, E3-1260L, E3-1220, E3-1230, E3-1240, E and E LINPACK Measurements with Pentium G620, Core i3-2100, i as well as Xeon E3-1220L, E3-1260L, E3-1220, E3-1230, E3-1240, E and E Version 1.1 Updated benchmark chapters: SPECcpu2006 Measurements with Celeron G530T and G530 STREAM Measurements with Celeron G530T and G530 LINPACK Measurements with Celeron G530T and G530 Version 1.2 Updated benchmark chapters: OLTP-2 New measurement results for Celeron G530 and G530T, Pentium G620 as well as Core i3 and Xeon E3 processor series Page 2 (24) Fujitsu Technology Solutions 2011

3 Technical data Like its predecessor the PRIMERGY RX100 S6, the PRIMERGY RX100 S7 is a mono socket rack server with one height unit. It includes an Intel C202 chip set, an Intel Celeron, Pentium, Core i3 or Xeon E3 processor, 4 DIMM slots for up to 32 GB DDR3-SDRAM, two onboard 1 GBit Ethernet controllers, a SATA- RAID controller and three PCI slots (1 PCI-Express 2.0 x16, 1 PCI-Express 2.0 x4 and 1 PCI-Express 2.0 x1). The PRIMERGY RX100 S7 comes in four versions: with a standard power supply unit and up to two 3.5 HDDs with a hot-plug power supply unit and up to two 3.5 HDDs with a standard power supply unit and up to four 2.5 HDDs or SSDs with a hot-plug power supply unit and up to four 2.5 HDDs or SSDs Detailed technical information is available in the data sheet PRIMERGY RX100 S7. Fujitsu Technology Solutions 2011 Page 3 (24)

4 SPECcpu2006 Benchmark description SPECcpu2006 is a benchmark which measures the system efficiency with integer and floating-point operations. It consists of an integer test suite (SPECint2006) containing 12 applications and a floating-point test suite (SPECfp2006) containing 17 applications. Both test suites are extremely computing-intensive and concentrate on the CPU and the memory. Other components, such as Disk I/O and network, are not measured by this benchmark. SPECcpu2006 is not tied to a special operating system. The benchmark is available as source code and is compiled before the actual measurement. The used compiler version and their optimization settings also affect the measurement result. SPECcpu2006 contains two different performance measurement methods: the first method (SPECint2006 or SPECfp2006) determines the time which is required to process single task. The second method (SPECint_rate2006 or SPECfp_rate2006) determines the throughput, i.e. the number of tasks that can be handled in parallel. Both methods are also divided into two measurement runs, "base" and "peak" which differ in the use of compiler optimization. When publishing the results the base values are always used; the peak values are optional. Benchmark Arithmetics Type Compiler optimization SPECint2006 integer peak aggressive SPECint_base2006 integer base conservative SPECint_rate2006 integer peak aggressive SPECint_rate_base2006 integer base conservative SPECfp2006 floating point peak aggressive SPECfp_base2006 floating point base conservative SPECfp_rate2006 floating point peak aggressive SPECfp_rate_base2006 floating point base conservative Measurement result Speed Throughput Speed Throughput Application single-threaded multi-threaded single-threaded multi-threaded The measurement results are the geometric average from normalized ratio values which have been determined for individual benchmarks. The geometric average - in contrast to the arithmetic average - means that there is a weighting in favour of the lower individual results. Normalized means that the measurement is how fast is the test system compared to a reference system. Value 1 was defined for the SPECint_base2006-, SPECint_rate_base2006, SPECfp_base2006 and SPECfp_rate_base2006 results of the reference system. For example, a SPECint_base2006 value of 2 means that the measuring system has handled this benchmark twice as fast as the reference system. A SPECfp_rate_base2006 value of 4 means that the measuring system has handled this benchmark some 4/[# base copies] times faster than the reference system. "# base copies specify how many parallel instances of the benchmark have been executed. Not every SPECcpu2006 measurement is submitted by for publication at SPEC. This is why the SPEC web pages do not have every result. As we archive the log files for all measurements, we can prove the correct implementation of the measurements at any time. Benchmark results The PRIMERGY RX100 S7 was measured with four different processor versions: Celeron Pentium Core i3 Xeon E3 The benchmark programs were compiled with Intel C++/Fortran Compiler 12 and run under SUSE Linux Enterprise Server 11 SP1 (64-bit). All the results in bold type in the following tables have been published at Page 4 (24) Fujitsu Technology Solutions 2011

5 Processor Cores GHz L3-Cache timing TDP SPECint_base2006 SPECint2006 Celeron G530T MB 1067 MHz 35 Watt Celeron G MB 1067 MHz 65 Watt Pentium G MB 1067 MHz 65 Watt Core i MB 1333 MHz 65 Watt Core i MB 1333 MHz 65 Watt Xeon E3-1220L MB 1333 MHz 20 Watt Xeon E3-1260L MB 1333 MHz 45 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 95 Watt Processor Cores GHz L3-Cache timing TDP SPECint_rate_base2006 SPECint_rate2006 Celeron G530T MB 1067 MHz 35 Watt Celeron G MB 1067 MHz 65 Watt Pentium G MB 1067 MHz 65 Watt Core i MB 1333 MHz 65 Watt Core i MB 1333 MHz 65 Watt Xeon E3-1220L MB 1333 MHz 20 Watt Xeon E3-1260L MB 1333 MHz 45 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 95 Watt Processor Cores GHz L3-Cache timing TDP SPECfp_base2006 SPECfp2006 Celeron G530T MB 1067 MHz 35 Watt Celeron G MB 1067 MHz 65 Watt Pentium G MB 1067 MHz 65 Watt Core i MB 1333 MHz 65 Watt Core i MB 1333 MHz 65 Watt Xeon E3-1220L MB 1333 MHz 20 Watt Xeon E3-1260L MB 1333 MHz 45 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 95 Watt Fujitsu Technology Solutions 2011 Page 5 (24)

6 Processor Cores GHz L3-Cache timing TDP SPECfp_rate_base2006 SPECfp_rate2006 Celeron G530T MB 1067 MHz 35 Watt Celeron G MB 1067 MHz 65 Watt Pentium G MB 1067 MHz 65 Watt Core i MB 1333 MHz 65 Watt Core i MB 1333 MHz 65 Watt Xeon E3-1220L MB 1333 MHz 20 Watt Xeon E3-1260L MB 1333 MHz 45 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 80 Watt Xeon E MB 1333 MHz 95 Watt Benchmark environment All SPECcpu2006 measurements were made on a PRIMERGY RX100 S7 with the following hardware and software configuration: Hardware Model CPU Number of CPUs Primary cache Secondary cache Other cache Software Operating System PRIMERGY RX100 S7 Celeron G530T, G530 Pentium G620 Core i3-2100, i Xeon E3-1220L, E3-1260L, E3-1220, E3-1230, E3-1240, E3-1270, E Celeron G530T, G530 2 cores Pentium G620: 2 cores Core i3-2100, i3-2120: 2 cores Xeon E3-1220L: 2 cores all others: 4 cores 32 KB instruction + 32 KB data on chip, per core 256 KB on chip, per core Celeron G530T, G530: 2 MB (I+D) on chip, per chip Pentium G620: 3 MB (I+D) on chip, per chip Core i3-2100, i3-2120: 3 MB (I+D) on chip, per chip Xeon E3-1220L: 3 MB (I+D) on chip, per chip all others: 8 MB (I+D) on chip, per chip 2 4 GB PC E DDR3-SDRAM SUSE Linux Enterprise Server 11 SP1 (64-bit) Compilers Intel C++/Fortran Compiler 12 Some components may not be available in all countries or sales regions. Page 6 (24) Fujitsu Technology Solutions 2011

7 SPECjbb2005 Benchmark description SPECjbb2005 is a Java business benchmark that focuses on the performance of Java Server platforms. SPECjbb2005 is essentially a modernized SPECjbb2000. The main differences are: The transactions have become more complex in order to cover a greater functional scope. The working set of the benchmark has been enlarged to the extent that the total system load has increased. SPECjbb2000 allows only one active Java Virtual Machine instance (JVM) whereas SPECjbb2005 permits several instances, which in turn achieves greater closeness to reality, particularly with large systems. On the software side SPECjbb2005 primarily measures the performance of the JVM used with its just-in-time compiler as well as their thread and garbage collection implementation. Some aspects of the operating system used also play a role. As far as hardware is concerned, it measures the efficiency of the CPUs and caches, the memory subsystem and the scalability of shared memory systems (SMP). Disk and network I/O are irrelevant. SPECjbb2005 emulates a 3-tier client/server system that is typical for modern business process applications with the emphasis on the middle-tier system: Clients generate the load, consisting of driver threads, which on the basis of TPC-C benchmark generate OLTP accesses to a database without thinking times. The middle tier system implements the business processes and the updating of the database. The database takes on the data management and is emulated by Java objects that are in the memory. Transaction logging is implemented on an XML basis. The major advantage of this benchmark is that it includes all three tiers that run together on a single host. The performance of the middle-tier is measured. Large-scale hardware installations are thus avoided and direct comparisons between the SPECjbb2005 results from the various systems are possible. Client and database emulation are also written in Java. SPECjbb2005 only needs the operating system as well as a Java Virtual Machine with J2SE 5.0 features. The scaling unit is a warehouse with approx. 25 MB Java objects. Precisely one Java thread per warehouse executes the operations on these objects. The business operations are assumed by TPC-C: New Order Entry Payment Order Status Inquiry Delivery Stock Level Supervision Customer Report However, these are the only features SPECjbb2005 and TPC-C have in common. The results of the two benchmarks are not comparable. SPECjbb2005 has 2 performance metrics: bops (business operations per second) is the overall rate of all business operations performed per second. bops/jvm is the ratio of the first metrics and the number of active JVM instances. In comparisons of various SPECjbb2005 results, both metrics must be specified. The following rules, according to which a compliant benchmark run has to be performed, are the basis for these three metrics: A compliant benchmark run consists of a sequence of measuring points with an increasing number of warehouses (and thus of threads) with the number in each case being increased by one warehouse. The run is started at one warehouse up through 2*MaxWh, but not less than 8 warehouses. MaxWh is the number of warehouses with the highest rate per second the benchmark expects. Per default the benchmark equates MaxWh with the number of CPUs visible by the operating system. The metric bops is the arithmetic average of all measured operation rates with MaxWh warehouses up to 2*MaxWh warehouses. Fujitsu Technology Solutions 2011 Page 7 (24)

8 Benchmark results In April 2011 the PRIMERGY RX100 S7 with the Xeon E processor and 32 GB of PC E DDR3-SDRAM memory was measured. The measurement was taken under Windows Server 2008 R2 Enterprise with SP1. As JVM a single instance of the Java HotSpot(TM) 64-Bit Server VM on Windows, version 1.6.0_25 from Oracle was used. The following result was obtained: SPECjbb2005 bops = SPECjbb2005 bops/jvm = The PRIMERGY RX100 S7 achieved the best result of all Intel-based 1-socket servers. 1 The following graphics illustrate the throughput of the PRIMERGY RX100 S7 in comparison to its predecessor PRIMERGY RX100 S6, in the respective most performant configuration. SPECjbb2005 bops: PRIMERGY RX100 S7 vs. predecessor SPECjbb2005 bops: PRIMERGY RX100 S7 vs. predecessor Benchmark environment The SPECjbb2005 measurement was run on a PRIMERGY RX100 S7 with the following hardware and software configuration: Hardware Model PRIMERGY RX100 S7 Processor Xeon E Number of chips 1 chip, 4 cores, 4 cores per chip, 2 threads per core Primary Cache 32 KB instruction + 32 KB data on chip, per core Secondary Cache 256 KB (I+D) on chip, per core Tertiary Cache 8 MB (I+D) on chip, per chip 4 x 8 GB PC E DDR3-SDRAM Software Operating System Windows Server 2008 R2 Enterprise SP1 JVM Version Oracle Java HotSpot(TM) 64-Bit Server VM on Windows, version 1.6.0_25 Some components may not be available in all countries or sales regions. 1 The comparative values mentioned above reflect the status of May 6th, This comparison is based on the SPECjbb2005 results of 1-socket servers with Intel processors. For the latest SPECjbb2005 results, visit Page 8 (24) Fujitsu Technology Solutions 2011

9 SPECpower_ssj2008 Benchmark description SPECpower_ssj2008 is the first industry-standard SPEC benchmark that evaluates the power and performance characteristics of a server. With SPECpower_ssj2008 SPEC has defined standards for server power measurements in the same way they have done for performance. The benchmark workload represents typical server-side Java business applications. The workload is scalable, multi-threaded, portable across a wide range of platforms and easy to run. The benchmark tests CPUs, caches, the memory hierarchy and scalability of symmetric multiprocessor systems (SMPs), as well as the implementation of Java Virtual Machine (JVM), Just In Time (JIT) compilers, garbage collection, threads and some aspects of the operating system. SPECpower_ssj2008 reports power consumption for servers at different performance levels from 100% to active idle in 10% segments over a set period of time. The graduated workload recognizes the fact that processing loads and power consumption on servers vary substantially over the course of days or weeks. To compute a power-performance metric across all levels, measured transaction throughputs for each segment are added together and then divided by the sum of the average power consumed for each segment. The result is a figure of merit called overall ssj_ops/watt. This ratio provides information about the energy efficiency of the measured server. The defined measurement standard enables customers to compare it with other configurations and servers measured with SPECpower_ssj2008. The diagram shows a typical graph of a SPECpower_ssj2008 result. The benchmark runs on a wide variety of operating systems and hardware architectures and does not require extensive client or storage infrastructure. The minimum equipment for SPEC-compliant testing is two networked computers, plus a power analyzer and a temperature sensor. One computer is the System Under Test (SUT) which runs one of the supported operating systems and the JVM. The JVM provides the environment required to run the SPECpower_ssj2008 workload which is implemented in Java. The other computer is a Control and Collection System (CCS) which controls the operation of the benchmark and captures the power, performance and temperature readings for reporting. The diagram provides an overview of the basic structure of the benchmark configuration and the various components. Fujitsu Technology Solutions 2011 Page 9 (24)

10 Benchmark results In May 2011 the PRIMERGY RX100 S7 was measured with one Xeon E3-1260L processor and 8 GB of PC3L-10600E DDR3-SDRAM memory. The measurement was taken under Windows Server 2008 R2 Enterprise with Service Pack 1. Java HotSpot(TM) 64-Bit Server VM on Windows, version 1.6.0_25 from Oracle was used as JVM. The PRIMERGY RX100 S7 with the Xeon E3-1260L processor achieved a new record result of 4,323 overall ssj_ops/watt in the rack-server class and thus surpassed the previous front runner from HP in this class, the ProLiant DL170e G6, by 34.6%. 2 The PRIMERGY RX100 S7 thus proves itself to be the most energy-efficient rack-server worldwide. The adjoining diagram shows the result of the configuration described above, measured with the PRIMERGY RX100 S7. The red horizontal bars show the performance to power ratio in ssj_ops/watt (upper x-axis) for each target load level tagged on the y-axis of the diagram. The blue line shows the run of the curve for the average power consumption (bottom x-axis) at each target load level marked with a small rhomb. The black vertical line shows the benchmark result of 4,323 overall ssj_ops/ watt for the PRIMERGY RX100 S7. This is the quotient of the sum of the transaction throughputs for each load level and the sum of the average power consumed for each measurement interval. The following table shows the benchmark results for the throughput in ssj_ops, the power consumption in watts and the resulting energy efficiency for each load level. Performance Power Energy Efficiency Target Load ssj_ops Average Power (W) ssj_ops/watt 100% 319, ,590 90% 290, ,401 80% 257, ,324 70% 228, ,161 60% 192, ,853 50% 161, ,440 40% 129, ,886 30% 97, ,224 20% 64, ,426 10% 32, ,413 Active Idle ssj_ops / power = 4,323 2 The comparative values mentioned above reflect the status of June 15th, This comparison is based on the most energy-efficient rack-server results. For the latest SPECpower_ssj2008 benchmark results, visit Page 10 (24) Fujitsu Technology Solutions 2011

11 The configuration was tuned to get the best possible result for this server in terms of performance per watt. The memory configuration with 2 x 4 GB was selected to meet the criteria of best performance at lowest power consumption by populating only one slot of each available memory channel. This configuration enables the benchmark to use the full potential of the available memory bandwidth and at the same time consumes less power than a 4-DIMM configuration with similar throughput. However, the most important factor in the hardware configuration is the correct choice of the processor. Processors are the part of a server which consume the most power besides the memory subsystem. This result with the highest efficiency level was achieved for the PRIMERGY RX100 S7 with the low-voltage Quad-Core Xeon E3-1260L processor with Thermal Design Power (TDP) of 45 watts. The following diagrams show for each load level the power consumption (on the right y-axis) and the throughput (on the left y-axis) of the PRIMERGY RX100 S7 compared to the predecessor the PRIMERGY RX100 S6. SPECpower_ssj2008 PRIMERGY RX100 S7 vs. predecessor In their most energy-efficient configurations the PRIMERGY RX100 S7 and the PRIMERGY RX100 S6 have approximately the same throughput. However, the PRIMERGY RX100 S7 achieves this throughput with less than half the power consumption. This results in an overall increase in energy efficiency in the PRIMERGY RX100 S7 of 101%. SPECpower_ssj2008 overall ssj_ops/watt: PRIMERGY RX100 S7 vs. predecessor Fujitsu Technology Solutions 2011 Page 11 (24)

12 Benchmark environment The SPECpower_ssj2008 measurement presented here was performed on a PRIMERGY RX100 S7 with the following hardware and software configuration using the ZES Zimmer LMG95 power analyzer: Hardware Model PRIMERGY RX100 S7 Processor (TDP) Xeon E3-1260L (45 W) Number of chips Primary Cache 1 chip, 4 cores per chip, 2 threads per core 32 KB instruction + 32 KB data on chip, per core Secondary Cache 256 KB (I+D) on chip, per core Tertiary Cache Network Interface Disk Subsystem 8 MB (I+D) on chip, per chip 2 x 4 GB PC3L-10600E DDR3-SDRAM 1 x Intel 82579LM Gigabit Network Connection (onboard) 1 x Integrated SATA controller 1 x 2.5 SATA SSD, 64 GB, JBOD Power Supply Unit 1 x 300 W Fujitsu Technology Solutions S26113-E577-V70-01 Software Operating System Windows Server 2008 R2 Enterprise SP1 JVM Version JVM affinity JVM options Oracle Java HotSpot(TM) 64-Bit Server VM on Windows, version 1.6.0_25 None -server -Xmx6g -Xms6g -Xmn5g -XX:SurvivorRatio=55 -XX:TargetSurvivorRatio=90 -XX:ParallelGCThreads=8 -XX:AllocatePrefetchDistance=256 -XX:AllocatePrefetchLines=4 -XX:LoopUnrollLimit=45 -XX:InitialTenuringThreshold=12 -XX:MaxTenuringThreshold=15 -XX:InlineSmallCode=3900 -XX:MaxInlineSize=270 -XX:FreqInlineSize=2500 -XX:+UseLargePages -XX:+UseParallelOldGC -XX:+UseCompressedStrings -XX:+AggressiveOpts Some components may not be available in all countries / sales regions. Page 12 (24) Fujitsu Technology Solutions 2011

13 OLTP-2 Benchmark description OLTP stands for Online Transaction Processing. The OLTP-2 benchmark is based on the typical application scenario of a database solution. In OLTP-2 database access is simulated and the number of transactions achieved per second (tps) determined as the unit of measurement for the system. In contrast to benchmarks such as SPECint and TPC-E, which were standardized by independent bodies and for which adherence to the respective rules and regulations are monitored, OLTP-2 is an internal benchmark of Fujitsu. OLTP-2 is based on the well-known database benchmark TPC-E. OLTP-2 was designed in such a way that a wide range of configurations can be measured to present the scaling of a system with regard to the CPU and memory configuration. Even if the two benchmarks OLTP-2 and TPC-E simulate similar application scenarios using the same load profiles, the results cannot be compared or even treated as equal, as the two benchmarks use different methods to simulate user load. OLTP-2 values are typically similar to TPC-E values. A direct comparison, or even referring to the OLTP-2 result as TPC-E, is not permitted, especially because there is no priceperformance calculation. Further information can be found in the document Benchmark Overview OLTP-2. Fujitsu Technology Solutions 2011 Page 13 (24)

14 Benchmark results The OLTP-2 values for the Intel Celeron, Pentium, Core i3 and Xeon E3 processor series, which are used in the current 1-socket systems, were determined. The following table gives you an overview of the processors considered and their properties: Celeron Processor #Cores/ Chip L3 Cache Processor Frequency Bus HT TM TDP G530T 2 2 MB 2.00 GHz 1066 MHz W G MB 2.40 GHz 1066 MHz W Pentium G MB 2.60 GHz 1066 MHz W Core i3 i MB 3.10 GHz 1333 MHz - 65 W i MB 3.30 GHz 1333 MHz - 65 W E3-1220L 2 3 MB 2.00 GHz 1333 MHz 20 W E3-1260L 4 8 MB 2.40 GHz 1333 MHz 45 W Xeon E MB 3.10 GHz 1333 MHz - 80 W E MB 3.20 GHz 1333 MHz 80 W E MB 3.30 GHz 1333 MHz 80 W E MB 3.40 GHz 1333 MHz 80 W E MB 3.50 GHz 1333 MHz 95 W HT = Hyper-Threading, TM = Turbo Mode, TDP = Thermal Design Power Processor TX120 S3 TX140 S1 RX100 S7 Celeron G530T G530 Pentium G620 Core i3 i i E3-1220L E3-1260L E Xeon E E E E Database performance greatly depends on the configuration options with CPU, memory and on the connectivity of an adequate disk subsystem for the database. Here, the systems vary in their CPU configuration options, as can be seen in the table opposite. In the following scaling considerations we assume that the disk subsystem has been adequately chosen and is not a bottleneck. The OLTP-2 values determined are based on the operating system Microsoft Windows Server 2008 R2 Enterprise and the database SQL Server 2008 R2 Enterprise x64 Edition. A guideline in the database environment for selecting main memory is that sufficient quantity is more important than the speed of the memory accesses. For this reason the maximum configuration with 8 GB modules was considered. Page 14 (24) Fujitsu Technology Solutions 2011

15 The following diagram shows the OLTP-2 transaction rates that can be achieved with one processor of the processor types under review here. OLTP-2 tps bold, cursive number: measured result others: calculated result Xeon E Core Xeon E Core Xeon E Core Xeon E Core Xeon E Core Xeon E3-1260L 4 Core Xeon E3-1220L 2 Core Core i Core Core i Core Pentium G620 2 Core Celeron G530 2 Core Celeron G530T 2 Core tps It is evident that a wide performance range is covered by the variety of released processors. If you compare the OLTP-2 value of the processor with the lowest performance (Celeron G530T) and the processor with the highest performance (Xeon E3-1280), the result is a 4.5-fold increase in performance. Based on the results achieved and the technical features the processors can be divided into different performance groups: Celeron and Pentium as the processors with two cores only and without Hyper-Threading make the start. The next performance group of processors achieves a higher performance in the OLTP-2 scenario. These are the processors with four cores and without turbo mode (Core i and Core i3-2120). In the group of Xeon E3 processors the Xeon E3-1220L with its two processor cores and an L3 cache of only 3 MB is at the lower end of the performance scale, but the power consumption with 20 W TDP is the smallest of all the processors under review here. The processors with four cores achieve a considerably better performance, because typically under the OLTP-2 load, doubling the number of cores almost results in twice the performance. Doubling the logical processor cores through Hyper-Threading also leads to better results under the OLTP-2 load, which explains the leap in performance of the Xeon E CPU without Hyper-Threading to the Xeon E processor with Hyper-Threading. Fujitsu Technology Solutions 2011 Page 15 (24)

16 Benchmark environment A typical OLTP-2 benchmark environment is shown symbolically in the following diagram: Driver Tier A Tier B Network Network Storage Subsystem Application Server Database Server Clients System under Test (SUT) All OLTP-2 values are based on the following configuration: Database Server (Tier B) Hardware System PRIMERGY RX100 S7 Processor 1 Celeron G530T (2C, 2.00 GHz) 1 Celeron G530 (2C, 2.40 GHz) 1 Pentium G620 (2C, 2.60 GHz) 1 Core i (4C, 3.10 GHz) 1 Core i (4C, 3.30 GHz) 1 Xeon E3-1220L (2C, 2.00 GHz) 1 Xeon E3-1260L (4C, 2.40 GHz) 1 Xeon E (4C, 3.10 GHz) 1 Xeon E (4C, 3.20 GHz) 1 Xeon E (4C, 3.30 GHz) 1 Xeon E (4C, 3.40 GHz) 1 Xeon E (4C, 3.50 GHz) 32 GB, 1333 MHz registered ECC DDR3 (8 GB DIMMs) Network interface 1 onboard LAN 1 Gb/s Disk subsystem RAID 0 (OS) Operating system and database application RAID 1 (LOG) Sequential access, optimized to short response times RAID 5 (data) Random access, optimized to throughput Software Operating system Windows Server 2008 R2 Enterprise Database SQL Server 2008 R2 Enterprise x64 Application Server (Tier A) Hardware System Processor Network interface Disk subsystem Software Operating system PRIMERGY RX200 S6 1 Xeon E5640 (6C, 2.66 GHz) 12 GB, 1333 MHz registered ECC DDR3 2 onboard LAN 1 Gb/s 1 73 GB 15k rpm SAS Drive Windows Server 2008 R2 Standard Page 16 (24) Fujitsu Technology Solutions 2011

17 Clients Hardware System PRIMERGY RX200 S5 Processor 2 Xeon X5570 (4C, 2.93 GHz) 24 GB, 1333 MHz registered ECC DDR3 Network interface 2 onboard LAN 1 Gb/s Disk subsystem 1 73 GB 15k rpm SAS Drive Software Operating system Windows Server 2008 R2 Standard OLTP-2 software EGen version Some components may not be available in all countries / sales regions. Fujitsu Technology Solutions 2011 Page 17 (24)

18 STREAM Benchmark description STREAM is a synthetic benchmark that has been used for many years to determine memory throughput and which was developed by John McCalpin during his professorship at the University of Delaware. Today STREAM is supported at the University of Virginia, where the source code can be downloaded in either Fortran or C. STREAM continues to play an important role in the HPC environment in particular. It is for example an integral part of the HPC Challenge benchmark suite. The benchmark is designed in such a way that it can be used both on PCs and on server systems. The unit of measurement of the benchmark is GB/s, i.e. the number of gigabytes that can be read and written per second. STREAM measures the memory throughput for sequential accesses. These can generally be performed more efficiently than accesses that are randomly distributed on the memory, because the CPU caches are used for sequential access. Before execution the source code is adapted to the environment to be measured. Therefore, the size of the data area must be at least four times larger than the total of all CPU caches so that these have as little influence as possible on the result. The OpenMP program library is used to enable selected parts of the program to be executed in parallel during the runtime of the benchmark, consequently achieving optimal load distribution to the available processor cores. During implementation the defined data area, consisting of 8-byte elements, is successively copied to four types, and arithmetic calculations are also performed to some extent. Type Execution Bytes per step Floating-point calculation per step COPY a(i) = b(i) 16 0 SCALE a(i) = q b(i) 16 1 SUM a(i) = b(i) + c(i) 24 1 TRIAD a(i) = b(i) + q c(i) 24 2 The throughput is output in GB/s for each type of calculation. The differences between the various values are usually only minor on modern systems. In general, only the determined TRIAD value is used as a comparison. The measured results primarily depend on the clock frequency of the memory modules; the CPUs influence the arithmetic calculations. The accuracy of the results is approximately 5%. This chapter specifies throughputs on a basis of 10 (1 GB/s = 10 9 Byte/s). Page 18 (24) Fujitsu Technology Solutions 2011

19 Benchmark results The PRIMERGY RX100 S7 was measured with four different processor versions: Celeron Pentium Core i3 Xeon E3 The benchmark was compiled using the Intel C compiler 12.0 and performed under SUSE Linux Enterprise Server 11 SP1 (64-bit). The data area consisted of 40 million elements, which is equivalent to about 305 MB. Processor Cores GHz L3 cache timing TDP TRIAD [GB/s] Celeron G530T MB 1067 MHz 35 Watt 15.4 Celeron G MB 1067 MHz 65 Watt 15.5 Pentium G MB 1067 MHz 65 Watt 15.6 Core i MB 1333 MHz 65 Watt 19.2 Core i MB 1333 MHz 65 Watt 19.2 Xeon E3-1220L MB 1333 MHz 20 Watt 19.0 Xeon E3-1260L MB 1333 MHz 45 Watt 18.7 Xeon E MB 1333 MHz 80 Watt 18.8 Xeon E MB 1333 MHz 80 Watt 18.8 Xeon E MB 1333 MHz 80 Watt 18.8 Xeon E MB 1333 MHz 80 Watt 18.8 Xeon E MB 1333 MHz 95 Watt 18.8 The results show the difference between the processors with a maximum memory frequency of 1067 MHz and those with 1333 MHz. Among the processors with a memory clock frequency of 1333 MHz those with 4 cores do not provide any better memory throughput than the ones with 2 cores, because the capacity limit of the memory controller is reached with 4 threads. The following diagram illustrates the throughput of the PRIMERGY RX100 S7 in comparison to its predecessor, the PRIMERGY RX100 S6, in its most performant configuration. GB/s STREAM: PRIMERGY RX100S7 vs. predecessor 16.5 PRIMERGY RX100 S6 Xeon X PRIMERGY RX100 S7 Xeon E STREAM TRIAD Fujitsu Technology Solutions 2011 Page 19 (24)

20 Benchmark environment All STREAM measurements were based on a PRIMERGY RX100 S7 with the following hardware and software configuration: Hardware Model CPU Number of cores Primary Cache Secondary Cache Other Cache Software Operating system PRIMERGY RX100 S7 Celeron G530T, G530 Pentium G620 Core i3-2100, i Xeon E3-1220L, E3-1260L, E3-1220, E3-1230, E3-1240, E3-1270, E Celeron G530T, G530: 2 cores Pentium G620: 2 cores Core i3-2100, i3-2120: 2 cores Xeon E3-1220L: 2 cores All others: 4 cores 32 KB instruction + 32 KB data on chip, per core 256 KB on chip, per core Celeron G530T, G530: 2 MB (I+D) on chip, per chip Pentium G620: 3 MB (I+D) on chip, per chip Core i3-2100, i3-2120: 3 MB (I+D) on chip, per chip Xeon E3-1220L: 3 MB (I+D) on chip, per chip All others: 8 MB (I+D) on chip, per chip 2 4 GB PC E DDR3-SDRAM SUSE Linux Enterprise Server 11 SP1 (64-bit) Compiler Intel C Compiler 12.0 Benchmark Stream.c Version 5.9 Some components may not be available in all countries or sales regions. Page 20 (24) Fujitsu Technology Solutions 2011

21 LINPACK Benchmark description LINPACK was developed in the 1970s by Jack Dongarra and some other people to show the performance of supercomputers. The benchmark consists of a collection of library functions for the analysis and solution of linear system of equations. A description can be found in the document LINPACK can be used to measure the speed of a computer during the solution of an N dimensional linear system of equations. The result is specified in GFlops (Giga Floating Point Operations per Second). It is a measure of how many floating-point operations can be carried out per second. The number of floating-point operations required for the solution is determined by the formula 2 / 3 N N 2. For the calculation LINPACK requires a matrix of size N N in the main memory with the value N standing for the number of equations to be solved. Maximum performance is achieved if the available main memory can be fully used as a result of choosing this value. However, the determination of this limit is very timeconsuming and the expected increase in the result is only minor. The memory bandwidth of the system also has hardly any impact on the result, because floating-point calculations are chiefly carried out during the run and data exchange only seldom takes place between the parallel processes. Thus the benchmark result is determined for a value of N that is somewhat below the maximum value. LINPACK is classed as one of the leading benchmarks in the field of high performance computing (HPC). LINPACK is one of the seven benchmarks currently included in the HPC Challenge benchmark suite, which takes other performance aspects in the HPC environment into account. An Intel-optimized LINPACK version for individual systems is used on PRIMERGY servers. This is part of the Intel Compiler or can also be directly downloaded from It is possible to publish LINPACK results at Prerequisite for this is the use of an MPIbased (Message Passing Interface) version. (See: The maximum theoretical performance of a processor core follows from the number of floating-point operations that are performed within a clock cycle. Thus e.g. a processor with a clocking frequency of 2.4 GHz and 4 floating-point operations per cycle would achieve a maximum performance of 9.6 GFlops. The ratio of the measured result to the maximum value shows the efficiency of the system for floating-point calculations. The fewer memory accesses required during the calculation, the better the ratio. Experience shows that current processor architectures achieve approximately 90%. Fujitsu Technology Solutions 2011 Page 21 (24)

22 Benchmark results The PRIMERGY RX100 S7 was measured with four different processor versions: Celeron Pentium Core i3 Xeon E3 The benchmark comes from the Intel Compiler 12.0 package and was performed under SUSE Linux Enterprise Server 11 SP1 (64-bit). The measured processors with the new AVX technology achieve 8 floating-point calculations per clock cycle. Thus, the theoretically achievable value is: GFlops max = 8 number of processor cores CPU frequency in GHz. The Pentium G620 and the Celeron processors with SSE4.2 technology only manage 4 floating-point calculations per clock cycle. The available main memory of 8 GB permits a dimension of N = Processor Cores GHz L3 cache [MB] timing TDP Theor. Max. [GFlops] LINPACK [GFlops] Efficiency [%] Celeron G530T MHz 35 Watt Celeron G MHz 65 Watt Pentium G MHz 65 Watt Core i MHz 65 Watt Core i MHz 65 Watt Xeon E3-1220L MHz 20 Watt Xeon E3-1260L MHz 45 Watt Xeon E MHz 80 Watt Xeon E MHz 80 Watt Xeon E MHz 80 Watt Xeon E MHz 80 Watt Xeon E MHz 95 Watt The results show that all the processors achieve more than 85% of the theoretical value and that the PRIMERGY TX120 S3 performs well in the field of floating-point calculations. Page 22 (24) Fujitsu Technology Solutions 2011

23 The following diagram illustrates the throughput of the PRIMERGY RX100 S7 in comparison to its predecessor, the PRIMERGY RX100 S6, in its most performant configuration. GFlops LINPACK: PRIMERGY RX100S7 vs. predecessor PRIMERGY RX100 S6 Xeon X3480 PRIMERGY RX100 S7 Xeon E Benchmark environment All LINPACK measurements were based on a PRIMERGY RX100 S7 with the following hardware and software configuration: Hardware Model CPU Number of cores Primary cache Secondary cache Other cache Software Operating system PRIMERGY RX100 S7 Celeron G530T, G530 Pentium G620 Core i3-2100, i Xeon E3-1220L, E3-1260L, E3-1220, E3-1230, E3-1240, E3-1270, E Celeron G530T, G530: 2 cores Pentium G620: 2 cores Core i3-2100, i3-2120: 2 cores Xeon E3-1220L: 2 cores All others: 4 cores 32 kb instruction + 32 kb data on chip, per core 256 kb on chip, per core Celeron G530T, G530: 2 MB (I+D) on chip, per chip Pentium G620: 3 MB (I+D) on chip, per chip Core i3-2100, i3-2120: 3 MB (I+D) on chip, per chip Xeon E3-1220L: 3 MB (I+D) on chip, per chip All others: 8 MB (I+D) on chip, per chip 2 4 GB PC E DDR3-SDRAM SUSE Linux Enterprise Server 11 SP1 (64-bit) Benchmark xlinpack_xeon64 from Intel Compiler 12.0 Some components may not be available in all countries or sales regions. Fujitsu Technology Solutions 2011 Page 23 (24)

24 Literature PRIMERGY Systems PRIMERGY RX100 S7 Data sheet RAID Controller Performance Single Disk Performance PRIMERGY Performance LINPACK OLTP-2 Benchmark Overview OLTP-2 SPECcpu Benchmark overview SPECcpu SPECjbb Benchmark overview SPECjbb SPECpower_ssj Benchmark Overview SPECpower_ssj STREAM Contact FUJITSU Technology Solutions Website: PRIMERGY Product Marketing PRIMERGY Performance and Benchmarks All rights reserved, including intellectual property rights. Technical data subject to modifications and delivery subject to availability. Any liability that the data and illustrations are complete, actual or correct is excluded. Designations may be trademarks and/or copyrights of the respective manufacturer, the use of which by third parties for their own purposes may infringe the rights of such owner. For further information see WW EN Copyright Fujitsu Technology Solutions GmbH 2011 Page 24 (24) Fujitsu Technology Solutions 2011