Analysis of K-bit pipelined processor cores using perl benchmarking

摘要

In today's high performance computing (HPC) environments, analyzing and predicting the performance of multiple-processor systems (clusters cores) on critical workloads remains a challenge. This is as a result of the key metrics that influences system's behavior. Busty arrivals in HPCs demand either a shared memory-parallel architecture or pipelined dataflow architecture. At present, a processor model that is deterministic while offering better bandwidth interconnection, excellent task allocation with the granularity of the task decomposition (overhead) is perceived as the most efficient type. This paper carried out a K-bit processor performance analysis with Perl Benchmark software platform. This is used in the evaluation of processor performance metrics. The software tool is developed with C# programming language and used to synthesize processor performance equations. Various processor cores are evaluated including Processor-1 (Pentium4), Processor-2(Core2Duo), Processor-3 (Core i7-Intel with 4-Cores and 8-Threads) and Processor-4 (AMD Opteron X4 2356). Direct measurement-form with a simulated/measured-form analysis is evaluated while analyzing the experimental test cases. The results indicate that the core i7 K-bit processor system offers the best performance considering execution time of 0.51secs (i.e. lower overhead) with 50% performance throughput hits relative to other CPU types. A clock cycle per instruction (CPI) of 0.60 is also obtained for core i7 showing the best latency profile. This observation can be used to determine optimal processor types for any given bandwidth and task decomposition operation.