Automation support for software performance engineering
Published 1 June 2001
Hesham El‐Sayed, Don Cameron, Murray Woodside
Citations16
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
An automated heuristic procedure for configuring a software system described by a layered architectural software model, onto a set of processors, and choosing priorities is described, which attempts to meet a soft-real-time performance specification.
Abstract
To evaluate the performance of a software design one must create a model of the software, together with the execution platform and configuration. Assuming that the "platform": (processors, networks, and operating systems) are specified by the designer, a good "configuration" (the allocation of tasks to processors, priorities, and other aspects of the installation) must be determined. Finding one may be a barrier to rapid evaluation; it is a more serious barrier if there are many platforms to be considered. This paper describes an automated heuristic procedure for configuring a software system described by a layered architectural software model, onto a set of processors, and choosing priorities. The procedure attempts to meet a soft-real-time performance specification, in which any number of scenarios have deadlines which must be realized some percentage of the time. It has been successful in configuring large systems with both soft and hard deadlines.
Keywords
Computer Science
Journal of the ACMScheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment
8,334 Citations1973C. L. Liu, J. W. Layland
University of Maribor digital library (University of Maribor)The Unified Modeling Language User Guide
5,529 Citations1999Grady Booch, James Rumbaugh +1 more
In The Unified Modeling Language User Guide, the original developers of the UML provide a tutorial to the core aspects of the language in a two-color format designed to facilitate learning.
The rate monotonic scheduling algorithm: exact characterization and average case behavior
1,461 Citations2003John P. Lehoczky, Lui Sha +1 more
An exact characterization of the ability of the rate monotonic scheduling algorithm to meet the deadlines of a periodic task set and a stochastic analysis which gives the probability distribution of the breakdown utilization of randomly generated task sets are represented.
SIAM Journal on ComputingAn Application of Bin-Packing to Multiprocessor Scheduling
657 Citations1978E. G. Coffman, M. R. Garey +1 more
This work considers one of the basic, well-studied problems of scheduling theory, that of nonpreemptively scheduling n independent tasks on m identical, parallel processors with the objective of minimizing the number of overlapping tasks.
Operations ResearchOn a Real-Time Scheduling Problem
567 Citations1978Sudarshan Dhall, C. L. Liu
This work studies the problem of scheduling periodic-time-critical tasks on multiprocessor computing systems and considers two heuristic algorithms that are easy to implement and yield a number of processors that is reasonably close to the minimum number.
IEEE Transactions on Software EngineeringThe Method of Layers
375 Citations1995Jerome Rolia, K. C. Sevcik
The Method of Layers (MOL) is proposed to provide performance estimates for distributed applications that contain one or more layers of software servers and uses the mean value analysis (MVA) linearizer algorithm as a subprogram to assist in predicting model performance measures.
Real-Time SystemsAllocating hard real-time tasks: An NP-Hard problem made easy
368 Citations1992K. Tindell, Alan Burns +1 more
An approach to solving the task allocation problem using a technique known as simulated annealing is described and a distributed hard real-time architecture is defined and new analysis is presented which enables timing requirements to be guaranteed.
IEEE Transactions on ComputersThe stochastic rendezvous network model for performance of synchronous client-server-like distributed software
302 Citations1995Murray Woodside, J. E. Neilson +2 more
Stochastic rendezvous networks are queueing networks of a new type which have been proposed as a modelling framework for these systems are extended to also incorporate different services or entries associated with each task, to give approximate performance estimates.
IEEE Transactions on Parallel and Distributed SystemsAllocation and scheduling of precedence-related periodic tasks
191 Citations1995Krithi Ramamritham
This paper discusses a static algorithm for allocating and scheduling components of periodic tasks across sites in distributed systems that handles precedence, communication, as well as replication requirements of subtasks of the tasks.
IEEE Transactions on ComputersTask Allocation and Precedence Relations for Distributed Real-Time Systems
166 Citations1987Chu Chu, Lan Lan
This paper presents a method for optimal module allocation that satisfies certain performance constraints and proposes an objective function that includes the intermodule communication (IMC) and accumulative execution time (AET) of each module.
IEEE Transactions on Software EngineeringAssignment and scheduling communicating periodic tasks in distributed real-time systems
158 Citations1997D.-T. Peng, Kang G. Shin +1 more
Presents an optimal solution to the problem of allocating communicating periodic tasks to heterogeneous processing nodes (PNs) in a distributed real-time system subject to the precedence constraints resulting from intercommunication among the tasks to be allocated.
Performance evaluation of software architectures
140 Citations1998Lloyd G. Williams, Connie U. Smith
The information required to perform such assessments is described and how it can be extracted from architectural descriptions is discussed and illustrated with a simple case study.
IEEE Transactions on Parallel and Distributed SystemsFast allocation of processes in distributed and parallel systems
122 Citations1993C.M. Woodside, G.G. Monforton
MULTIFIT-COM, a static task allocator which could be incorporated into an automated compiler/linker/loader for distributed processing systems, is presented and its clustering and load balancing properties on a large system are applied.
Optimized priority assignment for tasks and messages in distributed hard real-time systems
119 Citations2002J.J.G. Garcia, Michael González Harbour
A new heuristic algorithm for optimizing the assignment of priorities to tasks and messages in distributed hard realtime systems that executes two orders of magnitude faster than simulated annealing, finds better solutions, and finds solutions in cases where the latter method fails.
Performance EvaluationA toolset for performance engineering and software design of client-server systems
109 Citations1995Greg Franks, Alex Hubbard +5 more
TimeBench/SRVN is a prototype toolset for computer-aided design and performance analysis of software, with an emphasis on distributed client-server systems, and is intended as a demonstration of the concept of an integrated designer/performance interface, and as a research environment for fast analytic solvers for the models.
Performance EvaluationThroughput calculation for basic stochastic rendezvous networks
99 Citations1989C.M. Woodside
A distributed program is modelled as a stochastic network of tasks related by their rendezvous requests, which is assumed to have maximum concurrency as if each task were executed on its own processor.
IEEE Transactions on Software EngineeringModule allocation of real-time applications to distributed systems
56 Citations1990C.E. Houstics
Lecture notes in computer scienceUML-Based Performance Modeling Framework for Component-Based Distributed Systems
52 Citations2001Pekka Kähkipuro
A performance modeling framework that can be used in the development and maintenance of component-based distributed systems, such as those based on the CORBA, EJB, and COM+ platforms is described.
Allocation of periodic task modules with precedence and deadline constraints in distributed real-time systems
41 Citations2003Chao-Ju Hou, Kang G. Shin
The authors address the problem of allocating (assigning and scheduling) periodic task modules to processing nodes (PNs) in distributed real-time systems subject to task precedence and timing constraints using the branch-and-bound technique to find an optimal allocation that maximizes the probability of meeting task deadlines.
Real-Time SystemsA Heuristic Approach to the Multitask-Multiprocessor Assignment Problem using the Empty-Slots Method and Rate Monotonic Scheduling
30 Citations1997Jorge A. Santos, Edgardo Ferro +2 more
A heuristic approach to the problem of assigning a set of preemptible resource-sharing and blockable real-time tasks to be executed in aSet of heterogeneous processors communicated through an interprocessor network, which is NP-hard is presented.
Using UML models for performance calculation
22 Citations2000Fried Hoeben
This paper explores some ideas to automatically remove layers of indirection and abstraction that hinder straightforward performance evaluation of UML models by using them in a CASE tool.
Automated performance modeling from scenarios and SDL designs of distributed systems
16 Citations2002Hesham El‐Sayed, D. William Cameron +1 more
An automated approach is described for deriving a performance model directly from a software design of a complex system, in a way that is tightly coupled to the design process.
Heuristic algorithms for priority assignment in flow shops
4 Citations2002Reza Etemadi, G.M. Karam +1 more
Heuristic algorithms for priority assignment in a periodic flow shop model are introduced and their abilities to find a feasible schedule are compared to the existing heuristic algorithms by using simulation methods.
Fixed-priority scheduling of periodic tasks with end-to-end deadlines
2 Citations1997Jun Sun
The thesis describes five synchronization protocols, several deadline-based priority assignment methods, and corresponding schedulability analysis algorithms that form an integrated end-to-end scheduling framework for scheduling tasks with end- to-end deadlines and analyzing the Schedulability of an end-To-end system.