Dynamic Resource Management for Parallel Tasks in an Oversubscribed Energy-Constrained Heterogeneous Environment

Machovec, Dylan; Khemka, Bhavesh; Pasricha, Sudeep; Maciejewski, Anthony A.; Siegel, Howard Jay; Koenig, Gregory A.; Wright, Michael; Hilton, Marcia; Rambharos, Rejendra; Imam, Neena

doi:10.1109/ipdpsw.2016.25

Cited by 7 publications

(6 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This paper assumes that the system works under a system load, as a future work, we hope that we can propose some scheduling methods that which can adaptive to the change of the system load. As the energy-aware scheduling [37] is a hot topic in the scheduling and Cloudsim supports the DVFS model in the simulation, we would do some work which considers the energy consumption on Cloudsim.…”

Section: Discussionmentioning

confidence: 99%

A Scheduling Method of Moldable Parallel Tasks Considering Speedup and System Load on the Cloud

Zhang

2019

IEEE Access

View full text Add to dashboard Cite

The moldable parallel task (MPT) is a kind of parallel task that their sub-tasks hold the resources exclusively, which has been widely used in different areas. Our paper focuses on the scheduling of moldable tasks when every sub-task supports time-slice. The time-slice is a consecutive time that the sub-task holds the resources exclusively. After every time-slice, the sub-task can be canceled, suspended, or continued. We give the model of MPTs and propose a scheduling method for MPTs: MC (a heuristic scheduling method supporting time-slice model on the cloud). The simulation results show that, even under a forecast accuracy of system load under 90% and 95%, MC reduces average waiting time and average execution time; at the same time, MC has a lower value in the percentages of unfinished tasks. INDEX TERMS Moldable parallel task, speedup, parallelism, scheduling method.

show abstract

Section: Discussionmentioning

confidence: 99%

A Scheduling Method of Moldable Parallel Tasks Considering Speedup and System Load on the Cloud

Zhang

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…There is a large body of work on resource management strategies for HPC systems targeting various metrics such as system resource utilization, task execution time, and total power/energy consumption. Utility functions ( [12], [13], [17], [25]) have been shown to be an effective performance measure for resource management in an oversubscribed environment. Our work builds on this previous research to combine performance and energy to define VoS.…”

Section: Value Of Service For Resource Managementmentioning

confidence: 99%

“…A mapping event is when resource management decisions are made. In our environment, mapping events occur every minute, as typical at facilities such as those used by DoD [17]. At a mapping event, all tasks that have arrived in the system and have not started execution and have not been dropped are considered mappable since the last mapping event.…”

Section: Resource Management Heuristicsmentioning

confidence: 99%

“…In [13], heterogeneous cores are explicitly allocated. The energy aware utility based heuristics studied in ( [12], [17]) treat energy as a constraint instead of a component of the overall VoS. Utility maximization is also targeted in [25] under an energy constraint during oversubscription by prioritizing the scheduling of tasks with the highest magnitude of utility per unit energy at a particular instant of time.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Value-Based Resource Management in High-Performance Computing Systems

Machovec

Tunc

Kumbhare

et al. 2016

Proceedings of the ACM 7th Workshop on Scientific Cloud Computing

Self Cite

View full text Add to dashboard Cite

We introduce a new metric, Value of Service (VoS), which enables resource management techniques for high-performance computing (HPC) systems to take into consideration the value of completion time of a task and the value of energy used to compute that task at a given instant of time. These value functions have a soft-threshold, where the value function begins to decrease from its maximum value, and a hard-threshold, where the value function goes to zero. Each task has an associated importance factor to express the relative significance among tasks. We define the value of a task as the weighted sum of its value of performance and value of energy, multiplied by its importance factor. We also consider the variation in value for completing a task at different time; the value of energy reduction can change significantly between peak and non-peak periods. We define VoS for a given workload to be sum of the values for all tasks that are executed during a given period of time. Our system model is based on virtual machines (VMs), where each dynamically arriving task will be assigned to a VM with a resource configuration based on number of homogenous cores and amount of memory. Based on VoS, we design, evaluate, and compare different resource management heuristics. This comparison is done over various simulation scenarios and example experiments on an IBM blade server based system.

show abstract

“…Preliminary versions of portions of this material appear in the 2015 Metaheuristics International Conference [10] and the 2016 Heterogeneity in Computing Workshop [11]. The differences between this work and the preliminary versions include: (a) the design, analysis, and evaluation of two new metaheuristics that in general result in improved performance; (b) the simulation of many more environments, which is used to further analyze the performance of the heuristics and the effect that different parameters have on the heuristics; and (c) experiments on a physical testbed are used to further evaluate the heuristics.…”

Section: Introductionmentioning

confidence: 99%

Utility-based resource management in an oversubscribed energy-constrained heterogeneous environment executing parallel applications

et al. 2019

Self Cite

View full text Add to dashboard Cite

show abstract

Dynamic Resource Management for Parallel Tasks in an Oversubscribed Energy-Constrained Heterogeneous Environment

Cited by 7 publications

References 17 publications

A Scheduling Method of Moldable Parallel Tasks Considering Speedup and System Load on the Cloud

A Scheduling Method of Moldable Parallel Tasks Considering Speedup and System Load on the Cloud

Value-Based Resource Management in High-Performance Computing Systems

Utility-based resource management in an oversubscribed energy-constrained heterogeneous environment executing parallel applications

Contact Info

Product

Resources

About