Performance analysis using the MIPS R10000 performance counters
Marco Zagha, B. E. Larson, Steve Turner et al. · Proceedings - Supercomputing · 1996
: Tuning supercomputer application performance often requires analyzing the interaction of the application and the underlying architecture. In this paper, we describe support in the MIPS R10000 for non-intrusively monitoring a variety of processor events -- support that is particularly useful for characterizing the dynamic behavior of multi-level memory hierarchies, hardware-based cache coherence, and speculative execution. We first explain how performance data is collected using an integrated set of hardware mechanisms, operating system abstractions, and performance tools. We then describe several examples drawn from scientific applications, which illustrate how the counters and profiling tools provide information that helps developers analyze and tune applications. Keywords: performance analysis, profiling tools, hardware performance counters, MIPS R10000, SGI Power Challenge 1. Introduction A fundamental question asked by HPC application developers is: "Where is the time spent?"...
Cited by 210 works. doi:10.1145/369028.369059
Authors
References
- EEL
- Advanced performance features of the 64-bit PA-8000
- Embra
- PROTEUS: a high-performance parallel-architecture simulator
- Cache performance of the SPEC92 benchmark suite
- MINT: a front end for efficient simulation of shared-memory multiprocessors
- Characterizing the caching and synchronization performance of a multiprocessor operating system
- Cache profiling and the SPEC benchmarks: a case study
- ATOM: a flexible interface for building high performance program analysis tools
- Embra