Micro Benchmark Analysis of the KSRlO
نویسنده
چکیده
A new approach. micm benchmarks. has recently been developed. Using this technique. we have analyzed the KSRI, and in particular the "ALLCACHE" memory architecture and ring interconnection. We have been able to elucidate many facets of memory performance. The technique has enabled us to identify and characterize puts of the memory design not described by Kendall Square Research. Our results show that a miss in the local cache can incur a penally ranging hom 1.5 microseconds to 500 micmseconds (when a dirty "page" in the local cache must he evicted). The progranmer must be very careful in placement and accessing of data to obtain maximum performance from the KSRI: the data presented here will help in understanding the performance actually obtained.
منابع مشابه
Measuring Performance, Estimating Most Productive Scale Size, and Benchmarking of Hospitals Using DEA Approach: A Case Study in Iran
Background and Objectives: The goal of current study is to evaluate the performance of hospitals and their departments. This manuscript aimed at estimation of the most productive scale size (MPSS), returns to scale (RTS), and benchmarking for inefficient hospitals and their departments. Methods: The radial and non-radial data envelopment analysis (DEA) ap...
متن کاملFinding a suitable benchmark for commercial bank branches using DEA
This paper proposes a suitable benchmark for inefficient commercial bank branches by using Data Envelopment Analysis (DEA). In order to render an inefficient bank branch efficient, it is necessary to decrease inputs and increase outputs. As there are priorities for decreasing some certain inputs and increasing some certain outputs over other inputs and outputs, respectively, it is necessary ...
متن کاملBenchmark Forecasting in Data Envelopment Analysis for Decision Making Units
Although DEA is a powerful method in evaluating DMUs, it does have some limitations. One of the limitations of this method is the result of the evaluation is based on previously data and the results are not proper for forecasting the future changes. So For this purpose, we design feedback loops for forecasting inputs and outputs through system dynamics and simulation. Then we use DEA model to f...
متن کاملRANKING DMUS ON THE BENCHMARK LINE WITH EQUAL SHADOW PRICES
Data envelopment analysis (DEA) with considering the best condition for each decision making unit (DMU) assesses the relative efficiency for it and divides a homogenous group of DMUs in to two categories: efficient and inefficient, but traditional DEA models can not rank efficient DMUs. Although some models were introduced for ranking efficient DMUs, Franklin Lio & Hsuan peng (2008), proposed a...
متن کاملTechniques for Benchmarking of CPU Micro-Architecture for Performance Evaluation
CPU Micro-architecture has a significant impact on performance and hence is of special importance to the HighPerformance Computing industry. In this paper, we describe the development of a suite of benchmark programs which enables us to evaluate and compare processors based on the performance of the micro-architecture itself, independent of workload characteristics. The benchmark suite is compr...
متن کامل