Member Login

E-mail:    Password:  




 TitleDate AddedCompany
whitepaper Scalability and Performance of the HP ProLiant DL145 G2 Server With AMD Opteron Processors Model 254 (2.8 GHz/1 MB) in an HP SBC Environment2005-12-01 01:00:02
  The HP ProLiant DL145 G2 server is optimized for the High Performance Computing (HPC) market. The industry-leading performance features of this server include AMD Opteron processors, up to 16 GB of two-way interleaved PC3200 DDR memory, an embedded dual-channel Gigabit NIC, full-length and low profile 64-bit/133 MHz PCI-X slots, and optional X16 PCI Express support. Opteron processors can improve the performance of existing 32-bit applications while providing support for 64-bit addressing, breaking the 4 GB limit inherent in the earlier x86 processors and enhancing the performance of memory-bound applications. These capabilities combine with 1u-size and low power consumption to provide HP Server Based Computing (SBC) customers with a high-performance, low-cost, density-optimized server solution.   
whitepaper HP Brings Integrity to Washington2005-09-01 03:00:03
  Availability, reliability, manageability and scalability aren't new to the federal information technology (IT) lexicon. However, the ability to wrap these features into a consolidated, standards-based system only recently became a reality. And while many companies are still scrambling to provide solutions that meet emerging requirements, one has emerged that does so in a cost-effective manner: the HP Integrity server platform based on the Intel Itanium 2 processor.   
whitepaper Performance Analysis of Intel Pentium M and Pentium 4 Processors2005-09-15 03:00:04 Dell
  This white paper compares the architecture and performance of the Pentium M to higher frequency Pentium 4 processors. The study presented in this paper highlights the fact that, when evaluating mobile platform performance, factors other than processor frequency are becoming more important. The study shows that the Pentium M is competitive on almost all benchmarks. From a purely performance standpoint, its best showing is on the gaming benchmarks where it outperformed the Pentium 4 on several tests.   
whitepaper Linpack HPL Performance on IBM eServer 326 and xSeries 336 Servers2005-09-15 03:00:04
  This paper examines the performance of the Linpack benchmark on IBM eServer 326 (e326) and IBM eServer xSeries 336 (x336) systems. The e326 is powered by AMD Opteron processors, and the x336 is powered by Intel Xeon processors. In this paper, system factors are considered that affect benchmark performance, such as processor and network performance, and the paper shows that it is not sensitive to memory performance. The paper also considers ways to tune the benchmark. In particular the paper examines what are good values for P, Q, N, and NB, and concludes that optimal values for NB must be determined through experimentation, although many values of NB can be eliminated through simple analysis.   
whitepaper A Comparison of Single-Core and Dual-Core Opteron Processor Performance for HPC2005-07-01 IBM
  Dual-core AMD Opteron processors represent the latest significant development in microprocessor technology. This paper, using the IBM eServer 326, examines the performance of dual-core Opteron processors. Dual-core Opteron processors perform very well on applications that are highly parallel and whose performance is limited by the processor core, that is to say, by the frequency of the processor clock.

Tags: High Performance Computing		   
whitepaper Introducing Windows Server x64 on IBM eServer xSeries Servers2005-12-01 01:00:02
  This paper introduces Windows Server 2003, x64 Edition, the 64-bit processors from Intel and AMD that support it, and the xSeries servers on which the operating system runs. It describes the concepts of 64-bit computing and the technology behind the processors. This paper introduces the xSeries product line that includes the Intel EM64T processor. It also describes how the new Windows operating system is different from its 32-bit counterpart and discusses the migration options that are available to customers and developers.   
whitepaper Performance Scalability of Data-Mining Workloads in Bioinformatics2006-01-04 01:01:52
  The bioinformatics workloads studied in this paper are representative of general-purpose data-mining techniques. They use BNs, SVMs, and DP, among other methods. They are data parallel workloads with large data sets and are very compute intensive. Data mining is the extraction of hidden predictive information from large data bases. Emerging data mining applications are important factors to drive the architecture of future microprocessors. This paper analyzes the performance scalability on parallel architectures of such applications to understand how to best architect the next generation of microprocessors that will have many CPU cores on chip.   
whitepaper FPGAs for High-Performance DSP Applications2005-07-21 03:00:01
  This white paper compares the performance of DSP applications in Altera FPGAs with popular DSP processors as well as competitive FPGA offerings. With higher performance, you can easily time-division-multiplex your DSP design to increase the number of processing channels, reducing the overall cost of your system. The performance comparisons in this white paper use DSP IP benchmarks and application level benchmarks.   
whitepaper Energy Based Analysis of Embedded Microprocessor Cache Design2005-05-12 University of California
  This paper focuses on performing a broad exploration of the embedded processor cache design space in order to determine the energy consumption and performance characteristics of alternative cache designs. Using simulators and energy modeling frameworks, they seek to study different alternatives and identify the parameters critical for reducing energy consumption in caches. Based on the design space exploration, they identify cache size as the primary candidate for energy reduction. They explore different design alternatives that seek to reduce cache size without significantly affecting performance. They perform a re-analysis of a seminal work in low energy cache design which seeks to use reduction in cache size to curb energy consumption.

Tags: Embedded Microprocessors		   
whitepaper Performance of Voltaire InfiniBand in IBM 64-Bit Commodity HPC Clusters2005-09-15 03:00:04
  In this paper, the performance of Voltaire InfiniBand using 100 MHz PCI-X, 133 MHz PCI-X and PCI-Express slots has been examined. In this experimentation it is found that InfiniBand exhibits high throughput (1,735 MB/s) and low latency (4 microseconds) when used with PCI-Express. With 133 MHz PCI-X it also shows low latency (5 microseconds) and high bandwidth (788 MB/s), considering the limitations of its slot. InfiniBand shows consistent performance over a range of message sizes.