GitHub / open-power-sdk / cpi-breakdown

CPI refers to how many processor cycles are needed to complete an instruction. An instruction can be a read/write from memory operation, an arithmetic calculation, or bit-wise operation. The more cycles the processor takes to complete an instruction, the poorer the performance of the application in the processor. Application performance can be improved by decreasing the number of cycles that are needed for the processor to complete instructions. In the CPI breakdown model, a set of processor events is broken down into components. Processor performance counters calculate metrics for the event components. This approach provides a complete view of how the application behaves concerning processor performance.

JSON API: https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/open-power-sdk%2Fcpi-breakdown

Topics: cpi, cpi-breakdown

Files

Dependencies