The word performance in computer performance means “How well is the computer doing the work it is supposed to do?”. The average time it takes to render a screen or page. –What computer architecture –Why it is important to study –Organization and anatomy of computers –Impact of microelectronics technology on computers –Evolution and generations of the computer industry •This Lecture –Cost considerations in computer design –Why measuring performance is important –Different performance metrics Principles of scalable performance: 2 Performance metrics & measures, Speedup performance laws, Scalability analysis & approaches, Parallel Computer Architectures: 4 Design issues: communication models, interconnection networks, performance. Online application performance. It basically depends on response time, throughput and execution time of a computer … The basic characteristis of a computer system that we typically need to measure are: I a count of how many times an event occurs, I the duration of some time interval, and Geometric mean is better here (arithmetic mean can vary depending on the reference system). Partitioning and Scheduling of Program, Flow Mechanisms of a Program. Program Flow Mechanisms, System Interconnect Architectures, Principles of Scalable Performance, Performance. In general, the metrics are related to speed, accuracy, reliability and availability of services. Operational metrics. Metrics and Measures, Parallel Processing Applications, Speedup Performance Laws, Scalability Analysis and Approaches. Computer performance is the amount of work accomplished by a computer system. The most widely used metric is the performance on the SPEC benchmark Fall 2019 CS5513 Computer Architecture 8 Measuring Performance cont’d Normalized performance – Execution times are normalized to the performance of a reference system. System Interconnect Architectures, Scalable Performance Metrics, and Measures. Usually measures performance gains/losses over the reference system √ n ∏ i=0 From the Publisher: This book deals with advanced computer architecture and parallel programming techniques. Parallel Computer Models, Program and Network Properties, Parallelism Conditions. Advanced Computer Architecture Lab EECS Department, University of Michigan {blj,tnm}@umich.edu Abstract This report explains what it means to characterize the performance of a computer , and which methods are appro-priate and inappropriate for the task. Computer Organization & Architecture Moazzam Ali Sahi Computer Performance Metrics, Measurement, & Performance Metrics and Scalability Analysis October 10 – 11, 2002, Par.Comp Workshop at IIT-Delhi 2 v Requirements in performance and cost v Performance metrics v Work load and speed metrics v Communication overhead measurement Analysis v Theoretical concepts on Performance and Scalability • Phase Parallel Model The material is suitable for use as a textbook in a one-semester graduate or senior course,offered by Computer Science,Computer Engineering,Electrical Engineering,or Industrial Engineering programs. Introduction to Theory of Parallelism. View Week7_Lecture1.pdf from CS MISC at COMSATS Institute of Information Technology, Lahore. SIMD computers: array processors & … Performance Metric Metrics are criteria to compare the performances of a system.