Towards a hierarchical scheduling system for distributed WWW server clusters

Andresen, Daniel; McCune, Timothy

doi:10.1109/hpdc.1998.710016

Cited by 12 publications

(11 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We differ from many of these approaches [14,1,6,24,10,17,19,12,29] in that we are developing an autonomous scheduling strategy that does not require centralized control or knowledge for scheduling.…”

Section: Introduction 2 Related Workmentioning

confidence: 99%

A-FAST: Autonomous Flow Approach to Scheduling Tasks

Nandy

Carter

Ferrante

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. This paper investigates the problem of autonomously allocating a large number of independent, equal sized tasks on a distributed heterogeneous grid-like platform, using only local information. We propose A-FAST (Autonomous Flow Approach to Scheduling Tasks), an efficient, scalable, dynamic and generic (imposing no restrictions on the topology) protocol for this purpose. Motivated by the idea of pressure guiding the flow in fluid networks, A-FAST only uses parameters available locally to a node to guide scheduling decisions. Simulations show that the protocol performs well over a variety of networks, averaging more than 99.5% of the optimal performance and outperforms related techniques like RID (Receiver Initiated Diffusion). We also show how a modified use of local information can improve the performance of an unreliable system. Preliminary results from implementing A-FAST on a small but real-life distributed system show the performance of our protocol to be near the maximum throughput of the system. Such a protocol has the potential to aid the efficient deployment of large, data intensive applications on very large or dynamically changing heterogeneous peerto-peer computing platforms.

show abstract

Section: Introduction 2 Related Workmentioning

confidence: 99%

A-FAST: Autonomous Flow Approach to Scheduling Tasks

Nandy

Carter

Ferrante

2004

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…While the pervasiveness of the world wide web illustrates one obvious example, almost all computational constituencies have come to expect that some form of network connectivity will be attached to all potentially useful resources. 1 Supported by D ARPA N66001-97-C-8531 and NSF ASC-9701333 2 Supported by Department of Defense Modernization Program (NAVO) 3 Supported by N P ACI (NSF) With ubiquitous connectivity comes the ability to choose between otherwise equivalent resources based on their perceived performance. For example, students sharing resources in the computer science department at UCSD frequently try to choose the most lightly loaded server for their activities.…”

Section: Introductionmentioning

confidence: 99%

The network weather service: a distributed resource performance forecasting service for metacomputing

Wolski

Spring

Hayes

1999

Future Generation Computer Systems

994

572

View full text Add to dashboard Cite

The goal of the Network Weather Service is to provide accurate forecasts of dynamically changing performance characteristics from a distributed set of metacomputing resources. Providing a ubiquitous service that can both track dynamic performance changes and remain stable in spite of them requires adaptive programming techniques, an architectural design that supports extensibility, and internal abstractions that can beimplemented e ciently and portably. In this paper, we describe the current implementation of the NWS for Unix and TCP/IP sockets and provide examples of its performance monitoring and forecasting capabilities.

show abstract

“…H-SWEB provides a means by which requests can be dynamically scheduled across clusters of servers, optimizing the use of client resources as well as the scattered server nodes [2]. H-SWEB uses a cluster of servers, and can dynamically schedule a request across a group of nodes and between clusters of nodes.…”

Section: H-swebmentioning

confidence: 99%

“…Mobile agents are "programs that can be dispatched from one computer and transported to a remote computer for execution" [10]. H-SWEB is a hierarchical scheduler for distributed HTTP server clusters [2]. H-SWEB allows dynamic scheduling of HTTP requests on resources distributed across the Internet utilizing a distributed set of servers, set up in hierarchical clusters.…”

Section: Introductionmentioning

confidence: 99%