An empirical study of the reliability of UNIX utilities
Communications of the ACM
Adaptive Control
Software performance testing based on workload characterization
WOSP '02 Proceedings of the 3rd international workshop on Software and performance
The WSLA Framework: Specifying and Monitoring Service Level Agreements for Web Services
Journal of Network and Systems Management
The Vision of Autonomic Computing
Computer
Introduction to Stochastic Search and Optimization
Introduction to Stochastic Search and Optimization
Early performance testing of distributed software applications
WOSP '04 Proceedings of the 4th international workshop on Software and performance
Automated Generation of Resource Configurations through Policies
POLICY '04 Proceedings of the Fifth IEEE International Workshop on Policies for Distributed Systems and Networks
Database tuning advisor for microsoft SQL server 2005: demo
Proceedings of the 2005 ACM SIGMOD international conference on Management of data
Theory, Volume 1, Queueing Systems
Theory, Volume 1, Queueing Systems
Solving the starting problem: device drivers as self-describing artifacts
Proceedings of the 1st ACM SIGOPS/EuroSys European Conference on Computer Systems 2006
Modeling next generation configuration management tools
LISA '06 Proceedings of the 20th conference on Large Installation System Administration
An Architectural Framework for the Design and Analysis of Autonomous Adaptive Systems
COMPSAC '07 Proceedings of the 31st Annual International Computer Software and Applications Conference - Volume 01
NESTOR: an architecture for network self-management and organization
IEEE Journal on Selected Areas in Communications
An Analysis of Language-Level Support for Self-Adaptive Software
ACM Transactions on Autonomous and Adaptive Systems (TAAS)
Hi-index | 0.00 |
Resource managers (RMs) often expose configuration parameters that have a significant impact on the performance of the systems they manage. Configuring RMs is challenging because it requires accurate estimates of performance for a large number of configuration settings and many workloads, which scales poorly if configuration assessment requires running performance benchmarks. We propose an approach to evaluating RM configurations called model fuzzing that combines measurement and simple models to provide accurate and scalable configuration evaluation. Based on model fuzzing, we develop a methodology for configuring RMs that considers multiple evaluation criteria (e.g., high throughput, low number of threads). Applying this methodology to the .NET thread pool, we find a configuration that increases throughput by 240% compared with the throughput of a poorly chosen configuration. Using model fuzzing reduces the computational requirements to configure the .NET thread pool from machine-years to machine-hours.