Clustering and server selection using passive monitoring

Andrews, Mark P.; Shepherd, Bruce; Srinivasan, Aravind; Winkler, P.; Zane, F.

doi:10.1109/infcom.2002.1019425

Cited by 41 publications

(39 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…If there has been no traffic between peers of interest passive approaches cannot provide precise information. A representative of such a passive approach is the Shared Passive Network Performance Discovery (SPAND) tool [15] proposed by Seshan et al A different approach has been proposed by Andrews et al [16] who propose to cluster networks with respect to passive measurements.…”

Section: Passive Measurementsmentioning

confidence: 99%

Rational Server Selection for Mobile Agents

Pils

Diepolder

2003

Distributed Applications and Interoperable Systems

View full text Add to dashboard Cite

Abstract. As mobile agents have the ability to operate autonomously and in a disconnected way they are considered to suit mobile computing environments. Mobile users can dispatch agents into the fixed network where the agents operate in the users behalf. Thus, in contrast to client/server interactions agents do not suffer from poor performing wireless access networks. In this paper the performance of mobile agents and client/server interactions are analysed with respect to heterogeneous networks and server resources. It is argued that without a certain knowledge of the available resources agents can hardly decide whether they should migrate or just apply client/server calls to access a remote service. To this end, it is proposed that agents should access server selection systems in order to plan their migration strategy. However, while server selection systems process agent requests the agents are waiting idle. Thus, access to server selection systems comes at a cost and therefore agents must be careful about it. To solve this decision problem an algorithm is proposed which estimates the benefits of accessing server selection systems. Finally, the decision algorithm is evaluated with the help of a simulation model.

show abstract

Section: Passive Measurementsmentioning

confidence: 99%

Rational Server Selection for Mobile Agents

Pils

Diepolder

2003

Distributed Applications and Interoperable Systems

View full text Add to dashboard Cite

show abstract

“…In [17] the authors point out that keeping track of clients is extremely difficult and expensive (p 1).…”

Section: Mirrored Contentmentioning

confidence: 99%

“…In [17] the authors present a solution that involves clustering clients and directing clients in clusters to most beneficial servers through a DNS front end. Measurements are made through passive means by examining TCP information to determine the distance of a client to the servers.…”

Section: Dns Levelmentioning

confidence: 99%

“…The basic idea behind load balancing is to provide multiple servers with identical content and distributing clients to servers in a fashion that avoids requiring a server to deal with too many clients at one time. Load balancing of Internet sites has become a critical field of study [17]. As the Internet hosts more and more large files such as video, software, and scientific data sets it is crucial that load balancing mechanisms account for long, persistent downloads.…”

Section: Introductionmentioning

confidence: 99%

“…Another common method involves rewriting URLs (Unique Resource Locators) to point to unloaded servers from the multi-server system [2]. Yet another common method involves clusters of servers with one or many points of entry to serve as dispatchers to back-end content servers [17].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

HTTP 1.2: Distributed HTTP for Load Balancing Server Systems

O'Daniel¹

View full text Add to dashboard Cite

Content hosted on the Internet must appear robust and reliable to clients relying on such content. As more clients come to rely on content from a source, that source can be subjected to high levels of load. There are a number of solutions, collectively called load balancers, which try to solve the load problem through various means. All of these solutions are workarounds for dealing with problems inherent in the medium by which content is served thereby limiting their effectiveness. HTTP, or Hypertext Transport Protocol, is the dominant mechanism behind hosting content on the Internet through websites. The entirety of the Internet has changed drastically over its history, with the invention of new protocols, distribution methods, and technological improvements. However, HTTP has undergone only three versions since its inception in 1991, and all three versions serve content as a text stream that cannot be interrupted to allow for load balancing decisions. We propose a solution that takes existing portions of HTTP, augments them, and includes some new features in order to increase usability and management of serving content over the Internet by allowing redirection of content instream. This in-stream redirection introduces a new step into the client-server connection where servers can make decisions while continuing to serve content to the client. Load balancing methods can then use the new version of HTTP to make better decisions when applied to multi-server systems making load balancing more robust, with more control over the client-server interaction.

show abstract

The State of CDNs Today and What AI‐Assisted CDN Means for the Future

Sarkar

Majumdar

2023

Evolving Networking Technologies

View full text Add to dashboard Cite

Clustering and server selection using passive monitoring

Cited by 41 publications

References 6 publications

Rational Server Selection for Mobile Agents

Rational Server Selection for Mobile Agents

HTTP 1.2: Distributed HTTP for Load Balancing Server Systems

The State of CDNs Today and What AI‐Assisted CDN Means for the Future

Contact Info

Product

Resources

About