diff options
Diffstat (limited to 'benchmarks')
| -rw-r--r-- | benchmarks/bonnie/DESCR | 12 | ||||
| -rw-r--r-- | benchmarks/heapsort/DESCR | 4 | ||||
| -rw-r--r-- | benchmarks/linpack-bench/DESCR | 4 | ||||
| -rw-r--r-- | benchmarks/netio/DESCR | 4 | ||||
| -rw-r--r-- | benchmarks/xengine/DESCR | 2 |
5 files changed, 13 insertions, 13 deletions
diff --git a/benchmarks/bonnie/DESCR b/benchmarks/bonnie/DESCR index b93807f2ffc..a08bfc07576 100644 --- a/benchmarks/bonnie/DESCR +++ b/benchmarks/bonnie/DESCR @@ -3,15 +3,15 @@ Bonnie: Filesystem Benchmark Program Bonnie tests the speed of file I/O using standard C library calls. It does reads and writes of blocks, testing for the limit of sustained data rate (usually limited by the drive or controller) and updates on -a file (better simulating normal operating conditions and quite dependent +a file (better simulating normal operating conditions and quite dependent on drive and OS optimisations). The per character read and write tests are generally limited by CPU speed only on current generation hardware. It takes some 35 SPECint92 to read or write a file at a rate of 1MB/s using getc() and putc(). The seek tests are dependent on the buffer cache size, since the fraction -of disk blocks that fits into the buffer cache will be found without any -disk operation and will contribute zero seek time readings. I.e. if the -buffer cache is 16MB and the Bonnie test file is 32MB in size, then the -seek time will come out as half its real value. The seek time includes -rotational delay, and will thus always come out higher than specified for +of disk blocks that fits into the buffer cache will be found without any +disk operation and will contribute zero seek time readings. I.e. if the +buffer cache is 16MB and the Bonnie test file is 32MB in size, then the +seek time will come out as half its real value. The seek time includes +rotational delay, and will thus always come out higher than specified for a drive. diff --git a/benchmarks/heapsort/DESCR b/benchmarks/heapsort/DESCR index 5b5e893272b..32ef5d9360d 100644 --- a/benchmarks/heapsort/DESCR +++ b/benchmarks/heapsort/DESCR @@ -1,11 +1,11 @@ Based on the Heap Sort code in 'Numerical Recipes in C' by William H. Press, Brian P. Flannery, Saul A. Teukolsky, and William T. Vetterling, -Cambridge University Press, 1990, ISBN 0-521-35465-X. +Cambridge University Press, 1990, ISBN 0-521-35465-X. The MIPS rating is based upon the program run time (runtime) for one iteration and a gcc 2.1 unoptimized (gcc -DUNIX) assembly dump count of instructions per iteration for a i486 machine (assuming 80386 code). This -is the reference used. +is the reference used. The maximum amount of memory allocated is based on the 'imax' variable in main(). Memory size = (2000*sizeof(long))*2^imax. imax is currently set to diff --git a/benchmarks/linpack-bench/DESCR b/benchmarks/linpack-bench/DESCR index 948be490455..193f9057afb 100644 --- a/benchmarks/linpack-bench/DESCR +++ b/benchmarks/linpack-bench/DESCR @@ -1,4 +1,4 @@ -Translated to C by Bonnie Toy 5/88 +Translated to C by Bonnie Toy 5/88 LINPACK is a collection of Fortran subroutines that analyze and solve linear equations and linear least-squares problems. The package solves @@ -7,7 +7,7 @@ symmetric positive definite, triangular, and tridiagonal square. In addition, the package computes the QR and singular value decompositions of rectangular matrices and applies them to least-squares problems. LINPACK uses column-oriented algorithms to increase efficiency by preserving -locality of reference. +locality of reference. LINPACK was designed for supercomputers in use in the 1970s and early 1980s. LINPACK has been largely superceded by LAPACK, which has been diff --git a/benchmarks/netio/DESCR b/benchmarks/netio/DESCR index 48325aa6595..6b3e3517250 100644 --- a/benchmarks/netio/DESCR +++ b/benchmarks/netio/DESCR @@ -3,6 +3,6 @@ It measures the net throughput of a network via NetBIOS and/or TCP/IP protocols (Unix and DOS only support TCP/IP) using various different packet sizes. -One instance has to run on one machine as a server process, another -instance is used on another machine to perform the benchmark. When +One instance has to run on one machine as a server process, another +instance is used on another machine to perform the benchmark. When executed without arguments, the program will explain its usage. diff --git a/benchmarks/xengine/DESCR b/benchmarks/xengine/DESCR index 7cade0de30c..53896b3648c 100644 --- a/benchmarks/xengine/DESCR +++ b/benchmarks/xengine/DESCR @@ -1,4 +1,4 @@ -`xengine' is a reciprocating engine for X. This program used as +`xengine' is a reciprocating engine for X. This program used as benchmark for X. See README and man page for details. |