A scalable Multi-Tier Task Assignment Policy with Minimum Excess Load

Jayasinghe, Malith; Tari, Zahir; Zeephongsekul, Panlop

doi:10.1109/iscc.2010.5546501

Cited by 6 publications

(4 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this policy, when a task finishes the designated time limit without completing its processing, it will be continued at the next server. Recently, MTTMEL (Multi-Tier Task assignment policy with Minimum Excess Load) [24] was proposed. This policy, divides the cluster into tiers where each tier is associated with a time limit and contain one or more servers.…”

Section: Related Workmentioning

confidence: 99%

Improving the performance for task assignment in distributed server systems by partitioning the large tasks

Semchedine

Bouallouche-Medjkoune

Aı̈ssani

2014

International Journal of Computer Mathematics

View full text Add to dashboard Cite

Task Assignment in distributed server systems focuses on the policy that assigns the tasks reached these systems in order to improve the response time. These tasks, generally, have the property that there is a tiny fraction (about 3%) of the large tasks that makes half (50%) of the total load. However, this property creates additional problems: the large tasks make the load difficult to balance among the servers, and the small tasks will be delayed by the large ones when they are in the same queue. In this paper, we propose a new policy for the Web clusters that we call Partitioning Large Tasks (PLT) and which deals with these problems mostly under a high traffic demand and a high variability of task sizes. PLT partitions each large task into fragments and assigns them to be processed in a parallel way and completing at the same time to improve the mean response time, and separates the small tasks from the large tasks to avoid being delayed. Performance tests show a significantly improvement in performance of PLT over the existing task assignment policies.

show abstract

Section: Related Workmentioning

confidence: 99%

Improving the performance for task assignment in distributed server systems by partitioning the large tasks

Semchedine

Bouallouche-Medjkoune

Aı̈ssani

2014

International Journal of Computer Mathematics

View full text Add to dashboard Cite

show abstract

“…The first category includes policies that assume that there is no prior knowledge about the task size. Task assignment by guessing size (TAGS) policy (Harchol-Balter 2002), task assignment based on prioritising traffic flows (TAPTF) policy (Broberg, Tari, and Zeephongsekul 2004) and multi-tier task assignment policy with minimum excess load policy (Jayasinghe, Tari, and Zeephongsekul 2010) are examples of these policies. Policies in the second category are based on the assumption that the task size is known a priori.…”

Section: Literature Reviewmentioning

confidence: 99%

Fuzzy simulation approach to multi-task scheduling in uncertain distributed server systems

Azadeh

Farahani

Moradi

et al. 2014

International Journal of Production Research

View full text Add to dashboard Cite

“…MTTMEL (Multi-Tier Task assignment policy with Minimum Excess Load): MTTMEL (Jayasinghe et al, 2010) consists of a dispatcher and multiples tiers of servers. Each tier has one or more servers associated with a time limit of the tier.…”

Section: Policies Assuming No Priori Knowledge Of Task Sizementioning

confidence: 99%

“…In Jayasinghe et al (2010), the performance analysis between MTTMEL, TAGS and Random shows that with a high number of servers MTTMEL outperforms TAGS and Random in the case of a low variability and a high system load. However, when the number of servers is reduced, the performance of MTTMEL deteriorates under the empirical values of the variability.…”

Section: Policies Assuming No Priori Knowledge Of Task Sizementioning

confidence: 99%