Reality-based optimization
Scott McFarling · 2003
Profile-based optimization has been studied extensively. Numerous papers and real systems have shown substantial improvements. However, most of these papers have been limited to either branch prediction or instruction cache performance. Also, most of these papers have looked at small applications with a limited number of testing and training scenarios. In this paper, we look at real use of large real-world desktop applications. We also assume memory consumption and disk performance are the primary metrics of interest. For this domain, we show that it is very difficult to get adequate coverage of large applications even with an extensive collection of training scenarios. We propose instead to augment traditional scenarios with data derived from real use. We show that this methodology allows us to reduce memory pressure by 29% and disk reads by 33% compared to traditional approaches.
Cited by 3 works. doi:10.1109/cgo.2003.1191533
Authors
References
- ATOM
- Reducing the cost of branches
- SPEC CPU2000: measuring CPU performance in the New Millennium
- Predicting conditional branch directions from previous runs of a program
- System support for automatic profiling and optimization
- Program restructuring for virtual memory
- Dynamo
- Achieving high instruction cache performance with an optimizing compiler
- Global register allocation at link time
- Procedure merging with instruction caches