Performance visualization of a distributed system: a case study
H. Jakiela · Computer · 1995
The case study describes the use of a simple visualization technique to help resolve difficult performance problems in a large distributed system. The event trace data used was collected on a large distributed system built to run a benchmark of a complex securities trading application. The system was designed to handle several hundred transactions per second with subsecond response times. The complete benchmark workload consisted of several different types of transactions. For simplicity, the examples show only two types of transactions, representing about half the total workload.>