A national trace collection and distribution resource

Abstract

Computer systems are modeled before construction to minimize errors and performance bottlenecks. A common modeling approach is to build software models of comptuer system components, and use realistic trace data as input. This methodology is commonly referred to as trace-driven simulation. Trace-driven simulation can be very accurate if both the system model and input trace data represent the system under test. The accuracy of the model is typically under the control of the researcher, but so far little or no trace data has been available that accurately represents current or future workloads. To address this issue, we describe the Brigham Young Univ. Address Collection Hardware (BACH) and introduce our national trace distribution center and trace collection facility (http://traces.byu.edu). We also illustrate the types of traces we collect and make available to others.