Using control theory to achieve service level objectives in performance management

DOI10.1023/A:1015350520175MaRDI QIDQ698468zbMATH OpenOpenAlexFDO

Authors M. C. Fu

Publication date 18 September 2002

Published in Real-Time Systems (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1023/a:1015350520175

zbMATH Keywords

identificationintegral control lawLotus Notesoscillationsperformance managementqueue lengthroot-locus analysissoftware systemsystem modeling

Mathematics Subject Classification ID

Performance evaluation, queueing, and scheduling in the context of computer systems (68M20) Control/observation systems involving computers (process control, etc.) (93C83)

Review text

A widely used approach to achieving service level objectives for a software system is to add a controller that manipulates the target system's tuning parameters. A methodology for designing such controllers for software systems that builds on classical control theory is described. The classical approach proceeds in two steps: system identification and controller design. In system identification, the mathematical models of the target system are constructed. In controller design, an analysis of the models leads to a controller that will achieve the service level objectives. An analysis of a closed-loop system using an integral control law with Lotus Notes as the target is reported. The objective is to maintain a reference queue length. Using root-locus analysis from control theory it is possible to predict the occurrence of controller-unclouded oscillations in the system's response. Such oscillations are undesirable since they increase variability, thereby resulting in a failure to meet the service level objective. This controller is implemented for a real Lotus Notes system, and a remarkable correspondence between the behaviour of the real system and the predictions of the analysis is observed.

Recommendations

• An online predictive control framework for designing self-managing computing systems
• MODELING A CONTROL SYSTEM FOR ORGANIZATIONAL PROCESSES
• Distributed quantitative QoS control based on control theory in web clusters
• On the use of feedback control in the design of computing system components
• scientific article; zbMATH DE number 1958165

Cited in

(5)

• Power and performance control of soft real-time web server clusters
• Estimating Multiclass Service Demand Distributions Using Markovian Arrival Processes
• DeepRT: predictable deep learning inference for cyber-physical systems
• Experimental evaluation of linear time-invariant models for feedback performance control in real-time systems
• Quantifying and suppressing the measurement disturbance in feedback controlled real-time systems

This page was built for publication: Using control theory to achieve service level objectives in performance management

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q698468)