Selective Reservation Strategies for Backfill Job Scheduling

Srinivasan, Srividya; Kettimuthu, Rajkumar; Subramani, V.; Sadayappan, P.

doi:10.1007/3-540-36180-4_4

Cited by 99 publications

(89 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…D. Feytelson and A. Weil offered a conservative variant of Backfill algorithm [11]. Further, various modifications have been created by B. Lawson and E. Smyrni [13], D. Perkovic and P. Keleher [14], S. Srinivasan [15]. The main drawback of these algorithms is the time lag during calculation, which is not acceptable for critical services at the time of failure.…”

Section: Discussionmentioning

confidence: 99%

Approaches to the optimization of the placement of service-oriented cloud applications in the software-defined infrastructure of the virtual data center

Bolodurina¹,

Parfenov²,

Haenssgen³

2017

Collection of Selected Papers of the III International Conference on Information Technology and Nanotechnology

View full text Add to dashboard Cite

Nowadays, we see a steady growth in the use of service-oriented cloud applications in modern business. However, there are some issues related to the placement of service-oriented cloud applications in the software-defined infrastructure of the virtual data center. The goal of optimization is to control the service-oriented cloud applications within data centers. The advantage of modern infrastructure virtualization is the possibility to use software-defined networks and software-defined data storages. However, the existing optimization of algorithmic solutions does not take into account the specifics of working with multiple class service-oriented cloud applications types. The paper describes the models which describe the basic structures of service-oriented cloud applications including: a level distribution model of the software-defined infrastructure with the technology of cloud applications containerization, a generalized model of a service-oriented cloud application, a model of virtualization of service-oriented cloud applications based on containers. Besides, we developed the efficient algorithm for optimizing the technology of containerization of cloud applications and services in the virtual data center (VDC) infrastructure. We propose an efficient algorithm for placing applications in the infrastructure of a VDC The optimization of the placement of serviceoriented cloud applications based on the VM template and containers with VDC disabilities infrastructure is reduced to packing in containers. Besides, we generalize the well-known heuristic and deterministic Karmakar-Karp's algorithms. During the experimental studies, we have found that the use of our algorithm enables to decrease the response time of cloud applications and services and, therefore, to increase the productivity of user requests processing and to reduce the number of refusals.

show abstract

Section: Discussionmentioning

confidence: 99%

Approaches to the optimization of the placement of service-oriented cloud applications in the software-defined infrastructure of the virtual data center

Bolodurina¹,

Parfenov²,

Haenssgen³

2017

Collection of Selected Papers of the III International Conference on Information Technology and Nanotechnology

View full text Add to dashboard Cite

show abstract

“…Some jobs therefore have a trend opposite that of the entire job stream average. More work by Srinivasan et al [12] [13] show that the performance difference between conservative and aggressive backfilling is actually in (size based) job groups. The dominating group dictates the overall job stream relative performance.…”

Section: Performancementioning

confidence: 99%

“…We validate the results by using group-wise performance evaluation. Previous studies [10] [11] [12] [13] show that group-wise evaluation reveals job favoritism/starvation and hence scheduler effectiveness.…”

Section: Introductionmentioning

confidence: 99%

Characteristics of Fairness Metrics and Their Effect on Perceived Scheduler Effectiveness

Ngubiri

Vliet

2010

International Journal of Computers and Applications

View full text Add to dashboard Cite

Parallel job schedulers are mostly evaluated using performance metrics. Deductions however can be misleading due to selective job starvation (unfairness). To choose a better scheduler, therefore, there is a need to compare schedulers for fairness as well. Performance and fairness, however, have mostly been studied independently. We examine characteristics of three approaches to fairness evaluation in parallel job scheduling. We examine how they represent job starvation and other aspects of discrimination. We show that the implied unfairness is not always starvation/discrimination in practice. We use simultaneous consideration of performance and fairness and compare deductions with scheduler effectiveness derived from group-wise performance evaluation. We observe that due to possible misrepresentation of starvation by fairness metrics, schedulers shown as superior may not be so in practice.

show abstract

“…Srinivasan et al [22] use job groups to get a deeper understanding of scheduler robustness while scheduling moldable jobs. Deeper comparative studies between Conservative and Aggressive backfilling were studied in [25] [24] by studying performance of job groups in the job stream.…”

Section: Related Workmentioning

confidence: 99%

“…Srinivassan et al [22] demonstrates how a deeper understanding of robustness for moldable jobs' schedulers can be obtained using size based groups. Job groups were also used in [24] [25] when comparing performance of Conservative and Aggressive backfilling. More studies by Feitelson [12] employed group-wise analysis for different job streams and measurement metrics to show underlying performance implications.…”

Section: Introductionmentioning

confidence: 99%

Group-Wise Performance Evaluation of Processor Co-allocation in Multi-cluster Systems

Ngubiri

Vliet

2008

Job Scheduling Strategies for Parallel Processing

View full text Add to dashboard Cite

Abstract. Performance evaluation in multi-cluster processor co-allocation -like in many other parallel job scheduling problems-is mostly done by computing the average metric value for the entire job stream. This does not give a comprehensive understanding of the relative performance of the different jobs grouped by their characteristics. It is however the characteristics that affect how easy/hard jobs are to schedule. We, therefore, do not get to understand scheduler performance at job type level. In this paper, we study the performance of multi-cluster processor co-allocation for different job groups grouped by their size, components and widest component. We study their relative performance, sensitivity to parameters and how their performance is affected by the heuristics used to break them up into components. We show that the widest component us characteristic that most affects job schedulability. We also show that to get better performance, jobs should be broken up in such a way that the width of the widest component is minimized.

show abstract

Selective Reservation Strategies for Backfill Job Scheduling

Cited by 99 publications

References 14 publications

Approaches to the optimization of the placement of service-oriented cloud applications in the software-defined infrastructure of the virtual data center

Approaches to the optimization of the placement of service-oriented cloud applications in the software-defined infrastructure of the virtual data center

Characteristics of Fairness Metrics and Their Effect on Perceived Scheduler Effectiveness

Group-Wise Performance Evaluation of Processor Co-allocation in Multi-cluster Systems

Contact Info

Product

Resources

About