Abstract
This project focuses on how to benchmark a SMB/CIFS storage with trace and replay methodology. Traces are used primarily by file system researchers in an attempt to understand, categorize, and generalize file system workloads. However, because such traces provide a detailed information about how a specific system is actually used, they should also be of interest to system administrators. The goal of this thesis is to produce a trace-driven synthetic workload and show that it is similar with the original workload for all practical purposes. This has been achieved by examining properties of the original workload such as the inter-arrival time and request length. Upon examination of the original workload, it is shown that our system regenerate a synthetic workload with the same properties. The synthetic workload is also been used on several SAMBA/CIFS servers with different hardwares, the variety of the results show the system can be used to benchmark the performance of a system