Table of Contents
In this chapter we will try to address common pitfalls users can hit with PerfCake.
In a virtual environment, the host might have some system level tweaks to optimize its performance. Among others, some of the operating system real timer ticks or PIT/HPET timers might be skipped. This can lead in a poor PerfCake time measurement resolution. Pay close attention to the time benchmark executed during PerfCake startup. This reveals the maximal performance PerfCake is able to measure. All faster systems will look like having an infinite performance to PerfCake. PerfCake will also start displaying warning when such a speed is achieved and the test results will be flawed.