Sample traces of three benchmark runs collected with the TraceFS tracing system.

Additional information about TraceFS can be found at the following URL: http://www.filesystems.org/project-tracefs.html

Less

This is a set of system-call traces collected for the Carnegie Mellon Online Coda Project. The traces cover 33 hosts running Mach. several different workloads are represented, including file server, workstation, and time-sharing systems. Less

I/O traces taken at the system-call level in 2000 and 2001 as part of a security research project. The traces cover thirteen computers used for software development by CS researchers. The longest trace (machine02) runs from June 2000 to August 2001.

The traces are further described in a README file. The traces are in the SEER format; a tool to read the traces can be found here.

Less

I/O traces taken at the system-call level, excluding reads and writes but including opens, closes and all other I/O calls, and including process-related calls such as fork and exec. The traces cover nine laptop computers used for software development by CS researchers, some with very heavy activity. The longest trace (norgay) extends from April 2, 1996, to January 28, 1997; the shortest (crockett) is October 11 to November 19, 1996. All traces contain gaps representing times when the machines were rebooted or the trace software failed. Activity by the superuser is generally omitted.

Additional information about these traces can be found here. A tool to read these traces is here. Less

Traces collected for research as part of I/O Behavior Research at University of Wisconsin. Less