Profiling-as-a-Service: adaptive scalable resource profiling for the cloud in the cloud

Authors:
Nima Kaviani;Eric Wohlstadter;Rodger Lea
Affiliations:
Department of Computer Science, University of British Columbia, Vancouver, B.C., Canada;Department of Computer Science, University of British Columbia, Vancouver, B.C., Canada;Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, B.C., Canada
Venue:
ICSOC'11 Proceedings of the 9th international conference on Service-Oriented Computing
Year:
2011

Citing 9
Cited 0

A framework for reducing the cost of instrumented code

Proceedings of the ACM SIGPLAN 2001 conference on Programming language design and implementation
The Paradyn Parallel Performance Measurement Tool

Computer
Profiling Java applications using code hotswapping and dynamic call graph revelation

WOSP '04 Proceedings of the 4th international workshop on Software and performance
Low-overhead memory leak detection using adaptive statistical profiling

ASPLOS XI Proceedings of the 11th international conference on Architectural support for programming languages and operating systems
Pin: building customized program analysis tools with dynamic instrumentation

Proceedings of the 2005 ACM SIGPLAN conference on Programming language design and implementation
SuperPin: Parallelizing Dynamic Instrumentation for Real-Time Performance

Proceedings of the International Symposium on Code Generation and Optimization
AjaxScope: a platform for remotely monitoring the client-side behavior of web 2.0 applications

Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles
Integrated and Composable Supervision of BPEL Processes

ICSOC '08 Proceedings of the 6th International Conference on Service-Oriented Computing
Performance comparison of middleware architectures for generating dynamic web content

Proceedings of the ACM/IFIP/USENIX 2003 International Conference on Middleware

Quantified Score

Hi-index	0.00

Visualization

Abstract

Runtime profiling of Web-based applications and services is an effective method to aid in the provisioning of required resources, for monitoring service-level objectives, and for detecting implementation defects. Unfortunately, it is difficult to obtain accurate profile data on live client workloads due to the high overhead of instrumentation. This paper describes a cloud-based profiling service for managing the tradeoffs between: (i) profiling accuracy, (ii) performance overhead, and (iii) costs incurred for cloud computing platform usage. We validate our cloud-based profiling service by applying it to an open-source e-commerce Web application.