CachePerf: A New Tool for Classifying Different Types of Cache Misses Correctly
Tongping Liu, Ph.D.
• A new tool for classifying different types of cache misses correctly
• A unified profiler that can correctly identify different types of cache misses without manual involvement, including previously known and 4 new types of cache misses, resulting in significant improvement in the performance of applications
• Reports serious issues only, saving users from manual efforts on trivial issues
• Provides helpful information (and root cause) to guide manual fixes, without requiring the special expertise for further diagnosis
• Imposes reasonable overhead while identifying all types of cache misses correctly
The cache plays a key role in determining the performance of applications, no matter for sequential or concurrent programs on homogeneous and heterogeneous architecture. Therefore, it is important to locate and differentiate cache misses accurately. This invention provides a first unified profiling tool–CachePerf–that can correctly and efficiently identify different types of cache misses while imposing reasonable overhead, differentiate issues of allocators from those of applications, and exclude minor issues without much performance impact.
Dr. Tongping Liu is an Assistant Professor in the Department of Electrical and Computer Engineering at the University of Massachusetts Amherst. His research interests include the security, performance, and reliability of software systems.
Available for Licensing and/or Sponsored Research
UMA 22-015
Patent Pending
|
