Improving small job response time for opportunistic scheduling

Ghare, Gaurav D.; Leutenegger, Scott T.

doi:10.1109/mascot.2000.876584

Cited by 4 publications

(12 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Routing all the communication through a central server creates a bottleneck at the server. In addition, Condor applies FCFS queue for the jobs submitted by the same user, which is an inefficient way of scheduling generally, and especially in grid environment where computational power is abundant [7]. Examples of other schedulers that employ just in time or opportunistic scheduling are Pegasus [8] and Traverna [9].…”

Section: Introductionmentioning

confidence: 99%

Workflow scheduling in computational grids: Opportunistic vs. planned

Aziz

El‐Rewini

2008

2008 IEEE/ACS International Conference on Computer Systems and Applications

View full text Add to dashboard Cite

One of the prime-most issues that is faced by the parallel systems in the past and is inherited by the grid systems of today, is the mapping and the execution of jobs on the available machines. Various scheduling approaches that were suggested in classical parallel systems literature are adopted for the grid systems with appropriate modifications. Although these modifications made them suitable for execution in grid environment, these approaches failed to deliver on the performance front. Dynamic availability of resources, data dependence among the tasks and communication latencies are some of the factors that negatively affected the outcome of these scheduling algorithms. Various grid management systems implemented their custom tailored versions of scheduling algorithms for task execution in the grid. These approaches suffer from various shortcomings such as inefficient utilization of the execution queues and/or failure to cater for the dependencies among the tasks in a job. In this paper, we will compare the two prevalent scheduling models in the grid; opportunistic and planned. Opportunistic model is used in DAGMan for scheduling workflow applications, which in-turn is an integral component of Condor and Condor-G. Planned scheduling, on the other hand, is used by the grid management architectures such as GrADS and MARS.

show abstract

Section: Introductionmentioning

confidence: 99%

Workflow scheduling in computational grids: Opportunistic vs. planned

Aziz

El‐Rewini

2008

2008 IEEE/ACS International Conference on Computer Systems and Applications

View full text Add to dashboard Cite

show abstract

“…Through simulation, we observed that the proposed scheduling policy performs an order of magnitude better than the approach used in Condor as well as two newly proposed scheduling policies in [1]. Furthermore, this improved performance can be achieved without loss of throughput and it results in a more interactive nature of the system thus increasing its appeal.…”

Section: Introductionmentioning

confidence: 89%

“…However, in recent days, the usage patterns of the HTC system is changing in that users are not only submitting large jobs to the system that is scheduled at quite times such as at night but also small jobs that return results as quickly as possible and run during the working hours. For example Leutenegger [1] observed that scientists run long production run experiments in one portion of the parameter space. Meanwhile, the scientists explore other portions of the parameter space.…”

Section: Introductionmentioning

confidence: 99%

“…We note that opportunistic scheduling has been extended to work on geographically distributed systems [3] and support limited forms of parallelism [6]. The focus of these past work has been to develop mechanisms to maximize throughput and user functionality while there exists little work on how best to improve the opportunistic scheduling approaches for the scenarios described in [1], which we are interested in.…”

Section: Introductionmentioning

confidence: 99%

“…We are primarily interested in the relative performance of the HTC system for the daytime workloads as described in [1]. We focus on this situation here as these situations are becoming quite common in HTC environments.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Job scheduling policy for high throughput computing environments

Abawajy

Ninth International Conference on Parallel and Distributed Systems, 2002. Proceedings.

View full text Add to dashboard Cite

In high throughput computing environments, an opportunistic scheduling policy is used for placement of batch jobs on idle workstations for execution. In this paper, we propose a new opportunistic scheduling policy with the aim of exploiting the idle resources provided by systems such as Condor in an opportunistic manner. We compared the performance of the proposed scheduling policy with three policies namely the scheduling policy used in Condor, job rotate and round robin policies through simulation. The results show that the proposed scheduling policy offers substantial performance improvements over the exiting approaches. Furthermore, this improved performance is achieved without loss of throughput and it results in a more interactive nature of the system thus increasing its appeal.

show abstract

Job Scheduling Policy for High Throughput Grid Computing

Abawajy

2005

Distributed and Parallel Computing

View full text Add to dashboard Cite

Improving small job response time for opportunistic scheduling

Cited by 4 publications

References 12 publications

Workflow scheduling in computational grids: Opportunistic vs. planned

Workflow scheduling in computational grids: Opportunistic vs. planned

Job scheduling policy for high throughput computing environments

Job Scheduling Policy for High Throughput Grid Computing

Contact Info

Product

Resources

About