Livelock Avoidance for Meta-Schedulers
نویسندگان
چکیده
Meta-scheduling, a process which allows a user to schedule a job across multiple sites, has a potential for livelock. Current systems avoid livelock by locking down resources at multiple sites and allowing a metascheduler to control the resources during the lock down period or by limiting job size to that which will fit on one site. The former approach leads to poor utilization; the later poses limitations on job size. This research uses BYU's Meta-scheduler (YMS) which allows jobs to be scheduled across multiple sites without the need for locking down the nodes. YMS avoids livelock through exponential back-off This research quantifies the potential for livelock, determines a suitable back-off period, and provides a structure upon which to test theoretical local. schedulers. The results show that livelock exists and, that a suitable exponential back-off not only avoids livelock but reduces the scheduling time for each job.
منابع مشابه
Double auction-inspired meta-scheduling of parallel applications on global grids
Meta-schedulers map jobs to computational resources that are part of a grid, such as clusters, that in turn have their own local job schedulers. Existing Grid meta-schedulers either target system-centric metrics, such as utilization and throughput, or prioritize applications based on utility metrics provided by the users. The system-centric approach gives less importance to users’ individual ut...
متن کاملNature Inspired Schedulers in Computational Grids
ARTICLE 1. Introduction Computational Grids (CGs) are a natural extension of classical distributed systems. CGs are expected to leverage unprecedented larger computing capacities by virtually joining together geographically distributed resources at large scale. To achieve this objective, the design of ecient Grid schedulers that map and allocate tasks and applications onto Grid resources is a k...
متن کاملReview on a Dynamic Scheduling Algorithm for Grid with Task Duplication
Scheduling of tasks in the grid environment is challenging compared to that in local computing service facilities. In the field of distributed and parallel computing grid computing plays a vital role for solving a wide range of large scale parallel applications. The main goal of task scheduling is to achieve high system throughput and to match the application with the available computing resour...
متن کاملCharacterization of Bandwidth-aware Meta-schedulers for Co-allocating Jobs in a Mini-grid∗
In this paper, we present a bandwidth-centric job communication model that captures the interaction and impact of simultaneously co-allocated jobs in a grid. We compare our dynamic model with previous research that utilizes a fixed execution time penalty for co-allocated jobs. We explore the interaction of simultaneously co-allocated jobs and the contention they often create in the network infr...
متن کاملCASP: a community-aware scheduling protocol
The existing resource and topology heterogeneity has divided the scheduling solutions into local schedulers and high-level schedulers (a.k.a. meta-schedulers). Although much work has been proposed to optimise job queue based scheduling, seldom has attention been put on the job sharing behaviours between decentralised distributed resource pools, which in turn raises a notable opportunity to expl...
متن کامل