PerfMon

Performance Analysis Monitor

The two key parameters of Performance Verification are latency and bandwidth. By latency we mean the time taken by data to reach from one point to another and bandwidth means amount of data transferred in a given time. These two parameters have to be measured and verified against an expected value for all paths of interest and this process is known as performance verification.

With the complexity of SoCs increasing, importance of performance verification grows in verification cycle. So far, there is no standard approach and different approaches are taken based on the individual. In many cases, calculation is done based on reporting using post processing where testcase developer has to print unique and appropriate messages. However this requires development of extra script, overhead in terms of time and effort.

eInfochips has developed a Performance Monitor to standardize the process. It has following features:

Highly Configurable

  • Multiple paths and multiple traffic patterns supported
  • Configurable units for latency (ns, us, ms) and bandwidth (KBps, MBps, GBps, Kbps, Mbps, Gbps)
  • Configuration methods convenient, uncomplicated
  • Configuration compliance check

Reporting

  • Measurement report depending upon user defined verbose level
  • Error reporting in case of incorrect resultv
  • Run time reporting of measurements
  • Trace file reporting for all the measurements

Extended Capabilities

  • Callback support to collect functional coverage
  • API support to query performance statistics on-the-fly simulation
  • Supports facility to enable/disable majority of PerfMon features.
  • Configuration support to program Setup and Hold Transactions for Counter-Base window
  • Supports tolerance configuration to waive expected variance in performance result
  • Supports performance tracker (trace file) per traffic type per interface
  • Cumulative bandwidth measurement for multiple traffic patterns on same interface/path
  • Uniformity comparison between different interfaces
  • Root Mean Square (RMS) value calculation for latency, bandwidth and error values
  • CSV based configuration for each monitor

Types of Measurements

  • Per transaction & Average latency
  • Bandwidth
  • Alternate bandwidth and latency window
  • Cumulative bandwidth measurement for multiple traffic patterns
  • Uniformity comparison between interfaces
  • Root mean square (RMS) value calculation for latency, bandwidth and error values

It provides quick and efficient way to verify Performance requirements in complex SoCs having multiple interfaces with different protocols. Having plug-and-play protocol independent entity to measure performance helps in easy integration on bunch of targeted data paths and with the simple method of configuration user can easily control measurement types and report generation.

In addition to accuracy, the performance monitor also helps in reducing the time and effort of performance verification.

datasheet-icon

Download Data Sheet

Looking for Performance Analysis Monitor?