Differential profiling
Paul E. McKenney
- 01 Mar 1999
Vol. 29, Iss: 3, pp 219-234
TL;DR: In this paper, data from profiles taken at different loads may be combined to help locate performance bottlenecks that are distributed widely throughout a large program or system, such as those due to inlined functions or macros.
read more
Abstract: Performance is an important aspect of software quality; in some real-time systems, poor performance can cause physical damage or even deaths. This paper describes how data from profiles taken at different loads may be combined to help locate performance bottlenecks that are distributed widely throughout a large program or system, such as those due to inlined functions or macros. This paper also describes how this technique may be used to pinpoint several types of performance bottlenecks in large programs running on shared-memory multiprocessors. In this environment, the critical bottleneck might consume only a small fraction of the total resources (due to Amdahl's law) and might be widely distributed throughout the program under test. Such a bottleneck can be very difficult to find when using traditional profiling techniques. >
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
•Proceedings Article
Proceedings of the Twenty-Fourth ACM Symposium on Operating Systems Principles
Michael Kaminsky,Mike Dahlin +1 more
- 03 Nov 2013
TL;DR: This year's SOSP program includes 30 papers, and touches on a wide range of computer systems topics, from kernels to big data, from responsiveness to correctness, and from devices to data centers.
110
Understanding PARSEC performance on contemporary CMPs
Major Bhadauria,Vincent M. Weaver,Sally A. McKee +2 more
- 04 Oct 2009
TL;DR: This work finds new Chip Multiprocessor (CMP) designs to be largely compute-bound, and thus limited by number of cores, micro-architectural resources, and cache-to-cache transfers, rather than by off-chip memory or system bus bandwidth.
ScaAnalyzer: a tool to identify memory scalability bottlenecks in parallel programs
TL;DR: ScaAnalyzer provides high-level, detailed guidance to programmers for scalability optimization and identifies scalability bottlenecks caused by poor memory access behaviors and provides optimization guidance that yields significant improvement in scalability.
53
A Performance Anomaly Detection and Analysis Framework for DBMS Development
TL;DR: A framework to manage performance anomalies after establishing a set of conditions for a problem to be considered an anomaly is described, which uses Statistical Process Control (SPC) charts to detect performance anomalies and differential profiling to identify their root causes.
44
Diagnosing Performance Variations by Comparing Multi-Level Execution Traces
Francois Doray,Michel Dagenais +1 more
TL;DR: This work introduces TraceCompare, a framework that automatically identifies differences between groups of executions of the same task at the user space and kernel levels and demonstrates the efficiency of the approach by presenting four case studies in which TraceCompare was used to uncover serious performance problems in enterprise and open source applications, without any prior knowledge of their codebase.
References
•Proceedings Article
The Art of Computer Systems Performance Analysis.
Raj Jain
- 01 Jan 1990
TL;DR: The authors' goal is always to offer you an assortment of cost-free ebooks too as aid resolve your troubles.
4K
•Book
Introduction to operations research
C. West Churchman,Russell L. Ackoff,E. Leonard Arnoff +2 more
- 01 Jan 1957
767