Preemptive job scheduling policy for distributively-owned workstation clusters

Cluster computing has come to prominence as a cost-effective parallel processing tool for solving many complex computational problems. In this paper, we propose a new timesharing opportunistic scheduling policy to support remote batch job executions over networked clusters to be used in conjunction with the Condor Up-Down scheduling algorithm. We show that timesharing approaches can be used in an opportunistic setting to improve both mean job slowdowns and mean response times with little or no throughput reduction. We also show that the proposed algorithm achieves significant improvement in job response time and slowdown as compared to exiting approaches and some recently proposed new approaches.

Job scheduling policy for high throughput grid computing

The growing computational power requirements of grand challenge applications has promoted the need for merging high throughput computing and grid computing principles to harness computational resources distributed across multiple organisations. This paper identifies the issues in resource management and scheduling in the emerging high throughput grid computing context. We also survey and study the performance of several space-sharing and time-sharing opportunistic scheduling policies that have been developed for high throughput computing.

Fault-tolerant dynamic job scheduling policy

In this paper, we propose a scalable and fault-tolerant job scheduling framework for grid computing. The proposed framework loosely couples a dynamic job scheduling approach with the hybrid replications approach to schedule jobs efficiently while at the same time providing fault-tolerance. The novelty of the proposed framework is that it uses passive replication approach under high system load and active replication approach under low system loads. The switch between these two replication methods is also done dynamically and transparently.

Autonomic job scheduling policy for grid computing

Autonomic middleware services will play an important role in the management of resources and distributed workloads in emerging distributed computing environments. In this paper, we address the problem of autonomic grid resource scheduling and propose a scheduling infrastructure that is capable of self-management in the face of dynamic behavior inherent to this kind of systems.

Dynamic parallel job scheduling in multi-cluster computing systems

Job scheduling is a complex problem, yet it is fundamental to sustaining and improving the performance of parallel processing systems. In this paper, we address an on-line parallel job scheduling problem in heterogeneous multi-cluster computing systems. We propose a new space-sharing scheduling policy and show that it performs substantially better than the conventional policies.

Adaptive hierarchical scheduling policy for enterprise grid computing systems

In an enterprise grid computing environments, users have access to multiple resources that may be distributed geographically. Thus, resource allocation and scheduling is a fundamental issue in achieving high performance on enterprise grid computing. Most of current job scheduling systems for enterprise grid computing provide batch queuing support and focused solely on the allocation of processors to jobs. However, since I/O is also a critical resource for many jobs, the allocation of processor and I/O resources must be coordinated to allow the system to operate most effectively. To this end, we present a hierarchical scheduling policy paying special attention to I/O and service-demands of parallel jobs in homogeneous and heterogeneous systems with background workload. The performance of the proposed scheduling policy is studied under various system and workload parameters through simulation. We also compare performance of the proposed policy with a static space–time sharing policy. The results show that the proposed policy performs substantially better than the static space–time sharing policy.

An efficient adaptive scheduling policy for high-performance computing

The advent of commodity-based high-performance clusters has raised parallel and distributed computing to a new level. However, in order to achieve the best possible performance improvements for large-scale computing problems as well as good resource utilization, efficient resource management and scheduling is required. This paper proposes a new two-level adaptive space-sharing scheduling policy for non-dedicated heterogeneous commodity-based high-performance clusters. Using trace-driven simulation, the performance of the proposed scheduling policy is compared with existing adaptive space-sharing policies. Results of the simulation show that the proposed policy performs substantially better than the existing policies.

Reliability-aware distributed computing scheduling policy

One of the primary issues associated with the efficient and effective utilization of distributed computing is resource management and scheduling. As distributed computing resource failure is a common occurrence, the issue of deploying support for integrated scheduling and fault-tolerant approaches becomes paramount importance. To this end, we propose a fault-tolerant dynamic scheduling policy that loosely couples dynamic job scheduling with job replication scheme such that jobs are efficiently and reliably executed. The novelty of the proposed algorithm is that it uses passive replication approach under high system load and active replication approach under low system loads. The switch between these two replication methods is also done dynamically and transparently. Performance evaluation of the proposed fault-tolerant scheduler and a comparison with similar fault-tolerant scheduling policy is presented and shown that the proposed policy performs better than the existing approach.

Robust parallel job scheduling infrastructure for service-oriented grid computing systems

Recent trends in grid computing development is moving towards a service-oriented architecture. With the momentum gaining for the service-oriented grid computing systems, the issue of deploying support for integrated scheduling and fault-tolerant approaches becomes paramount importance. To this end, we propose a scalable framework that loosely couples the dynamic job scheduling approach with the hybrid replications approach to schedule jobs efficiently while at the same time providing fault-tolerance. The novelty of the proposed framework is that it uses passive replication approach under high system load and active replication approach under low system loads. The switch between these two replication methods is also done dynamically and transparently.

Evaluating the EASY-backfill job scheduling of static workloads on clusters

This research aims at improving our understanding of backfilling job scheduling algorithms. The most frequently used algorithm, EASY-backfilling, was selected for a performance evaluation by scheduling static workloads of parallel jobs on a computer cluster. To achieve the aim, we have developed a batch job scheduler for Linux clusters, implemented several scheduling algorithms including ARCA and EASY-Backfilling, and carried out their performance evaluation by running well known MPI applications on a real cluster. Our performance evaluation carried out for EASY-Backfilling serves two purposes. First, the performance results obtained from our evaluation can be used to validate other researcherspsila results generated by simulation, and second, the methodology used in our evaluation has alleviated many problems existed in the simulations presented in the current literature.

GLOA : a new job scheduling algorithm for grid computing

The purpose of grid computing is to produce a virtual supercomputer by using free resources available through widespread networks such as the Internet. This resource distribution, changes in resource availability, and an unreliable communication infrastructure pose a major challenge for efficient resource allocation. Because of the geographical spread of resources and their distributed management, grid scheduling is considered to be a NP-complete problem. It has been shown that evolutionary algorithms offer good performance for grid scheduling. This article uses a new evaluation (distributed) algorithm inspired by the effect of leaders in social groups, the group leaders' optimization algorithm (GLOA), to solve the problem of scheduling independent tasks in a grid computing system. Simulation results comparing GLOA with several other evaluation algorithms show that GLOA produces shorter makespans.

Job strain exposures vs. stress-related workers’ compensation claims in Victoria, Australia: developing a public health response to job stress

We present a comparative analysis of patterns of exposure to job stressors and stress-related workers’ compensation (WC) claims to provide an evaluation of the adequacy of claims-driven policy and practice. We assessed job strain prevalence in a 2003 population-based survey of Victorian [Australia] workers and compared these results with stress-related WC statistics for the same year. Job strain prevalence was higher among females than males, and elevated among lower vs. higher occupational skill levels. In comparison, claims were higher among females than males, but primarily among higher skill-level workers. There was some congruence between exposure and WC claims patterns. Highly exposed groups in lower socio-economic positions were underrepresented in claims statistics, suggesting that the WC insurance perspective substantially underestimates the job stress problems for these groups. Thus to provide a sufficient evidence base for equitable policy and practice responses to this growing public health problem, exposure or health outcome data are needed as an essential complement to claims statistics.

Load balancing based concurrent execution of NAS parallel benchmarks with BYTE sequential benchmarks on a non-dedicated cluster

Dedicated clusters are becoming commonly used for high performance parallel processing. Computers of a non-dedicated cluster are often idle or lightly loaded. These under utilised computers can be employed to execute parallel applications. Thus, they have to be shared by parallel and sequential applications, which could lead to the improvement of their execution performance. There is a lack of experimental study showing the behaviour and performance of executing parallel and sequential applications concurrently on a non-dedicated cluster. We present the result of an experimental study into load balancing of a mixture of parallel and sequential applications on a non-dedicated cluster.

Parallel computing on clusters and enterprise grids : practice and experience

We assert that companies can make more money and research institutions can improve their performance if inexpensive clusters and enterprise grids are exploited. In this paper, we have demonstrated that our claim is valid by showing the study of how programming environments, tools and middleware could be used for the execution of parallel and sequential applications, multiple parallel applications executing simultaneously on a non-dedicated cluster, and parallel applications on an enterprise grid and that the execution performance was improved. For this purpose an execution environment, and parallel and sequential benchmark applications selected for, and used in, the experiments were characterised.

Pareto archived simulated annealing for single machine job shop scheduling with multiple objectives

In this paper, the single machine job shop scheduling problem is studied with the objectives of minimizing the tardiness and the material cost of jobs. The simultaneous consideration of these objectives is the multi-criteria optimization problem under study. A metaheuristic procedure based on simulated annealing is proposed to find the approximate Pareto optimal (non-dominated) solutions. The two objectives are combined in one composite utility function based on the decision maker’s interest in having a schedule with weighted combination. In view of the unknown nature of the weights for the defined objectives, a priori approach is applied to search for the non-dominated set of solutions based on the Pareto dominance. The obtained solutions set is presented to the decision maker to choose the best solution according to his preferences. The performance of the algorithm is evaluated in terms of the number of non-dominated schedules generated and the proximity of the obtained non-dominated front to the true Pareto front. Results show that the produced solutions do not differ significantly from the optimal solutions.