China 
Africa 
Explore chapters and articles related to this topic
Introduction to computer architecture
Published in Joseph D. Dumas, Computer Architecture, 2016
Perhaps the best known and most popular benchmark suites in recent years have been the ones developed by the Open Systems Group (OSG) of the Standard Performance Evaluation Corporation (SPEC), formerly known as the System Performance Evaluation Cooperative. The distinguishing feature of the SPEC benchmark suites is that they are composed of actual application code, not synthetic or artificial code (like Whetstones and Dhrystones) written just to exercise the system. The SPEC CPU benchmarks primarily evaluate the performance of the system processor and memory as well as the compiler; they are not very I/O intensive and make few demands of the operating system. SPEC CPU includes both an integer benchmark suite (SPECint) and a floating-point suite (SPECfp). Each suite consists of several programs (currently 19 floating-point applications written in Fortran, C, and C++ and 12 integer applications written in C and C++) with a result reported for each. The overall SPECfp or SPECint rating is a geometric mean of a given system’s performances on each application.
Simple, Yet Fast and Effective Two-Phase Method for Nurse Rostering
Published in American Journal of Mathematical and Management Sciences, 2020
F. Guessoum, S. Haddadi, E. Gattal
Now, we compare the six methods from the point of view of speed. Usually, comparing computing times is problematic since the computing environments used are different (see Table 7). Researchers are used to standardize computing times via specint values. The SPEC (Standard Performance Evaluation Corporation) website https://www.spec.org/cpu2006/results/cint2006.html provides an estimate of the execution speed of many modern computers. A Specint2006 value is attached to each computer; the higher the value, the faster the computer. Unfortunately, the machines indicated in Table 7 and used by the competing heuristics are not included in the website. However, we took the values associated with machines with similar specifications whose specint values are reported again in Table 7. From these, scaling factors are estimated and utilized for standardizing the computing times in Table 9.