The Million Instructions Per Second (MIPS) graph gives you a general idea of CPU speed in a measured use case. The basic use case for the MIPS analysis would be to check any low areas in the graph since they may reveal unoptimized features, for example, in terms of memory access (reads and writes). The higher the MIPS values, the better and the faster the CPU is in executing instructions.
A MIPS (Million Instructions Per Millisecond) graph is displayed in the Performance Counters view, when Instructions executed has been selected as one of the profiled counters. From PI Profiler version 2.2.1. onwards, CPU speed data is included when "Instructions executed" profiled. For files created with earlier PI Profiler versions, the dialog below is displayed during PIAnalyser file import:
Figure 1. MIPS Graph Generation
The processor speed information is needed for calculating the data for the MIPS graph. The rest of the data for the calculation is always included in the Instructions executed and CPU cycles counters.
The MIPS counter information is displayed in the graphs view and the legend table together with the other performance counters. Note, however, that for MIPS only Average (1/ms), Min (1/ms), and Max (1/ms) information is displayed in the legend table.
When the CPU speed data is included in the imported Profiler data file, a CPU clock rate counter and graph are also displayed in the Performance Counters View.