Clustering and prefetching techniques for network-based storage systems

Thakker, Dhawal; Mapp, Glenford

doi:10.1109/ants.2008.4937801

“…The simulation was done by using a discrete eventbased system written in C++. More details are given in [15]. The simulation results are then compared with results for our analytical model.…”

Section: Comparison Of Resultsmentioning

confidence: 97%

“…These results were used to develop techniques for prefetching and clustering using a high-performance network storage server [16,15]. It should be noted that the model is approximate as it depends on which state of the imaginary chain is used to calculate r. This is because the solution for r varies slightly depending on which state is used.…”

Section: Comparison Of Resultsmentioning

confidence: 98%

See 1 more Smart Citation

Exploring gate-limited analytical models for high-performance network storage servers

Mapp

¹

,

Thakker

²

,

Gemikonakli

³

2011

Journal of Computer and System Sciences

1

0

View full text Add to dashboard Cite

Gate-limited service is a type of service discipline found in queueing theory and can be used to describe a number of operational environments, for example, large transport systems such as buses, trains or taxis, etc. Recently, there has been the observation that such systems can also be used to describe interactive Internet Services which use a Client/Server interaction. In addition, new services of this genre are being developed for the local area. One such service is a Network Memory Server (NMS) being developed here at Middlesex University. Though there are several examples of real systems that can be modelled using gate-limited service, it is fair to say that the analytical models which have been developed for gate-limited systems have been difficult to use, requiring many iterations before practical results can be generated. In this paper, a detailed gate-limited bulk service queueing model based on Markov chains is explored and a numerical solution is demonstrated for simple scenarios. Quantitative results are presented and compared with a mathematical simulation. The analysis is used to develop an algorithm based on the concept of optimum operational points. The algorithm is then employed to build a high-performance server which is capable of balancing the need to prefetch for streaming applications while promptly satisfying demand misses. The algorithm is further tested using a systems simulation and then incorporated into an Experimental File System (EFS) which showed that the algorithm can be used in a real networking environment.

show abstract

“…These results were used to develop techniques for prefetching and clustering using a high-performance network storage server [16,15]. It should be noted that the model is approximate as it depends on which state of the imaginary chain is used to calculate r. This is because the solution for r varies slightly depending on which state is used.…”

Section: Comparison Of Resultsmentioning

confidence: 99%

Exploring Gate-Limited Analytical Models for High Performance Network Storage Servers

Mapp

¹

,

Thakker

²

,

Gemikonakli

³

2009

2009 Proceedings of 18th International Conference on Computer Communications and Networks

View full text Add to dashboard Cite

Gate-limited service is a type of service discipline found in queueing theory and can be used to describe a number of operational environments, for example, large transport systems such as buses, trains or taxis, etc. Recently, there has been the observation that such systems can also be used to describe interactive Internet Services which use a Client/Server interaction. In addition, new services of this genre are being developed for the local area. One such service is a Network Memory Server (NMS) being developed here at Middlesex University. Though there are several examples of real systems that can be modelled using gate-limited service, it is fair to say that the analytical models which have been developed for gate-limited systems have been difficult to use, requiring many iterations before practical results can be generated. In this paper, a detailed gate-limited bulk service queueing model based on Markov chains is explored and a numerical solution is demonstrated for simple scenarios. Quantitative results are presented and compared with a mathematical simulation. The analysis is used to develop an algorithm based on the concept of optimum operational points. The algorithm is then employed to build a high-performance server which is capable of balancing the need to prefetch for streaming applications while promptly satisfying demand misses. The algorithm is further tested using a systems simulation and then incorporated into an Experimental File System (EFS) which showed that the algorithm can be used in a real networking environment.

show abstract

“…The function of this layer is to act as a common interface between the service delivery framework and the underlying network architecture such as IP overlay network or new technologies such as Y-Comm [16] which divide the Internet into a Core network surrounded by Peripheral wireless networks.…”

Section: Fig 3 Envisioned Cloud-based Service Frameworkmentioning

confidence: 99%

On the Investigation of Cloud-Based Mobile Media Environments With Service-Populating and QoS-Aware Mechanisms

Sardis

¹

,

Mapp

²

,

Loo

³

et al. 2013

IEEE Trans. Multimedia

View full text Add to dashboard Cite

Abstract-Recent advances in mobile devices and network technologies have set new trends in the way we use computers and access networks. Cloud Computing, where processing and storage resources are residing on the network is one of these trends. The other is Mobile Computing, where mobile devices such as smartphones and tablets are believed to replace personal computers by combining network connectivity, mobility, and software functionality. In the future, these devices are expected to seamlessly switch between different network providers using vertical handover mechanisms in order to maintain network connectivity at all times. This will enable mobile devices to access Cloud Services without interruption as users move around. Using current service delivery models, mobile devices moving from one geographical location to another will keep accessing those services from the local Cloud of their previous network, which might lead to moving a large volume of data over the Internet backbone over long distances. This scenario highlights the fact that user mobility will result in more congestion on the Internet. This will degrade the Quality of Service and by extension, the Quality of Experience offered by the services in the Cloud and especially multimedia services that have very tight temporal constraints in terms of bandwidth and jitter. We believe that a different approach is required to manage resources more efficiently, while improving the Quality of Service and Quality of Experience of mobile media services. This paper introduces a novel concept of Cloud-Based Mobile Media Service Delivery in which services run on localised public Clouds and are capable of populating other public Clouds in different geographical locations depending on service demands and network status. Using an analytical framework, this paper argues that as the demand for specific services increases in a location, it might be more efficient to move those services closer to that location. This will prevent the Internet backbone from experiencing high traffic loads due to multimedia streams and will offer service providers an automated resource allocation and management mechanism for their services.

show abstract

Clustering and prefetching techniques for network-based storage systems

Cited by 5 publications

References 6 publications

Exploring gate-limited analytical models for high-performance network storage servers

Exploring gate-limited analytical models for high-performance network storage servers

Exploring Gate-Limited Analytical Models for High Performance Network Storage Servers

On the Investigation of Cloud-Based Mobile Media Environments With Service-Populating and QoS-Aware Mechanisms

Contact Info

Product

Resources

About