On Sun, 17 Mar 2002, Nadav Har'El wrote: > On Sun, Mar 17, 2002, guy keren wrote about "Re: pthreads question": > > > > On Sat, 16 Mar 2002, Malcolm Kavalsky wrote: > > > > > I attach a program benchmark.c that compares speed of memcpy versus data > > > transfer > > > over unix sockets. > > > > on my PC (AMD k-6 II 366MHz w/256MB RAM, kernel 2.2.20): > > > > [choo@simey ~]$ gcc -O2 benchmark.c > > [choo@simey ~]$ ./a.out > > Memcpy'ed 2000 blocks of size 1048576 in 38 seconds => 52 Mbytes/second > > Sent 2000 blocks of size 1048576 in 94 seconds over unixsocket => 21 > > Mbytes/second > > [choo@simey ~]$ > > Similarly, on Redhat 7.2 and kernel 2.4.7-10, Pentium III (Katmai) 500 MHz: > $ a.out > Memcpy'ed 2000 blocks of size 1048576 in 12 seconds => 166 Mbytes/second > Sent 2000 blocks of size 1048576 in 14 seconds over unix socket => 142 Mbytes/second
Just for all to know that Kavalsky's results are not unique: this is on a MDK8.1 system (glibc-2.2.4, kernel 2.4.17) with Athlon 1GHz and 133MHz SDRAM: Memcpy'ed 2000 blocks of size 1048576 in 7 seconds => 285 Mbytes/second Sent 2000 blocks of size 1048576 in 4 seconds over unix socket => 500 Mbytes/second (the 7s is sometimes 8s, and the 4s is sometimes 5s). So there might be some improvements in the kernel between 2.4.7 and 2.4.17. -- Matan Ziv-Av. [EMAIL PROTECTED] ================================================================= To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]