Value of service based resource management for large-scale computing systems

Cihan Tunc, Dylan Machovec, Nirmal Kumbhare, Ali Akoglu, Salim A Hariri, Bhavesh Khemka, Howard Jay Siegel

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Task scheduling for large-scale computing systems is a challenging problem. From the users perspective, the main concern is the performance of the submitted tasks, whereas, for the cloud service providers, reducing operation cost while providing the required service is critical. Therefore, it is important for task scheduling mechanisms to balance users’ performance requirements and energy efficiency because energy consumption is one of the major operational costs. We present a time dependent value of service (VoS) metric that will be maximized by the scheduling algorithm that take into consideration the arrival time of a task while evaluating the value functions for completing a task at a given time and the tasks energy consumption. We consider the variation in value for completing a task at different times such that the value of energy reduction can change significantly between peak and non-peak periods. To determine the value of a task completion, we use completion time and energy consumption with soft and hard thresholds. We define the VoS for a given workload to be the sum of the values for all tasks that are executed during a given period of time. Our system model is based on virtual machines, where each task will be assigned a resource configuration characterized by the number of the homogeneous cores and amount of memory. For the scheduling of each task submitted to our system, we use the estimated time to compute matrix and the estimated energy consumption matrix which are created using historical data. We design, evaluate, and compare our task scheduling methods to show that a significant improvement in energy consumption can be achieved when considering time-of-use dependent scheduling algorithms. The simulation results show that we improve the performance and the energy values up to 49% when compared to schedulers that do not consider the value functions. Similar to the simulation results, our experimental results from running our value based scheduling on an IBM blade server show up to 82% improvement in performance value, 110% improvement in energy value, and up to 77% improvement in VoS compared to schedulers that do not consider the value functions.

Original languageEnglish (US)
Pages (from-to)1-18
Number of pages18
JournalCluster Computing
DOIs
StateAccepted/In press - May 19 2017

Fingerprint

Energy utilization
Scheduling
Scheduling algorithms
Energy efficiency
Costs
Servers
Data storage equipment

Keywords

  • Energy efficient resource allocation
  • Performance metrics
  • Resource management
  • Task scheduling
  • Value of service
  • Virtual machines

ASJC Scopus subject areas

  • Software
  • Computer Networks and Communications

Cite this

Value of service based resource management for large-scale computing systems. / Tunc, Cihan; Machovec, Dylan; Kumbhare, Nirmal; Akoglu, Ali; Hariri, Salim A; Khemka, Bhavesh; Siegel, Howard Jay.

In: Cluster Computing, 19.05.2017, p. 1-18.

Research output: Contribution to journalArticle

Tunc, Cihan ; Machovec, Dylan ; Kumbhare, Nirmal ; Akoglu, Ali ; Hariri, Salim A ; Khemka, Bhavesh ; Siegel, Howard Jay. / Value of service based resource management for large-scale computing systems. In: Cluster Computing. 2017 ; pp. 1-18.
@article{cc8217d735214e12bc95903020f8dc53,
title = "Value of service based resource management for large-scale computing systems",
abstract = "Task scheduling for large-scale computing systems is a challenging problem. From the users perspective, the main concern is the performance of the submitted tasks, whereas, for the cloud service providers, reducing operation cost while providing the required service is critical. Therefore, it is important for task scheduling mechanisms to balance users’ performance requirements and energy efficiency because energy consumption is one of the major operational costs. We present a time dependent value of service (VoS) metric that will be maximized by the scheduling algorithm that take into consideration the arrival time of a task while evaluating the value functions for completing a task at a given time and the tasks energy consumption. We consider the variation in value for completing a task at different times such that the value of energy reduction can change significantly between peak and non-peak periods. To determine the value of a task completion, we use completion time and energy consumption with soft and hard thresholds. We define the VoS for a given workload to be the sum of the values for all tasks that are executed during a given period of time. Our system model is based on virtual machines, where each task will be assigned a resource configuration characterized by the number of the homogeneous cores and amount of memory. For the scheduling of each task submitted to our system, we use the estimated time to compute matrix and the estimated energy consumption matrix which are created using historical data. We design, evaluate, and compare our task scheduling methods to show that a significant improvement in energy consumption can be achieved when considering time-of-use dependent scheduling algorithms. The simulation results show that we improve the performance and the energy values up to 49{\%} when compared to schedulers that do not consider the value functions. Similar to the simulation results, our experimental results from running our value based scheduling on an IBM blade server show up to 82{\%} improvement in performance value, 110{\%} improvement in energy value, and up to 77{\%} improvement in VoS compared to schedulers that do not consider the value functions.",
keywords = "Energy efficient resource allocation, Performance metrics, Resource management, Task scheduling, Value of service, Virtual machines",
author = "Cihan Tunc and Dylan Machovec and Nirmal Kumbhare and Ali Akoglu and Hariri, {Salim A} and Bhavesh Khemka and Siegel, {Howard Jay}",
year = "2017",
month = "5",
day = "19",
doi = "10.1007/s10586-017-0901-9",
language = "English (US)",
pages = "1--18",
journal = "Cluster Computing",
issn = "1386-7857",
publisher = "Kluwer Academic Publishers",

}

TY - JOUR

T1 - Value of service based resource management for large-scale computing systems

AU - Tunc, Cihan

AU - Machovec, Dylan

AU - Kumbhare, Nirmal

AU - Akoglu, Ali

AU - Hariri, Salim A

AU - Khemka, Bhavesh

AU - Siegel, Howard Jay

PY - 2017/5/19

Y1 - 2017/5/19

N2 - Task scheduling for large-scale computing systems is a challenging problem. From the users perspective, the main concern is the performance of the submitted tasks, whereas, for the cloud service providers, reducing operation cost while providing the required service is critical. Therefore, it is important for task scheduling mechanisms to balance users’ performance requirements and energy efficiency because energy consumption is one of the major operational costs. We present a time dependent value of service (VoS) metric that will be maximized by the scheduling algorithm that take into consideration the arrival time of a task while evaluating the value functions for completing a task at a given time and the tasks energy consumption. We consider the variation in value for completing a task at different times such that the value of energy reduction can change significantly between peak and non-peak periods. To determine the value of a task completion, we use completion time and energy consumption with soft and hard thresholds. We define the VoS for a given workload to be the sum of the values for all tasks that are executed during a given period of time. Our system model is based on virtual machines, where each task will be assigned a resource configuration characterized by the number of the homogeneous cores and amount of memory. For the scheduling of each task submitted to our system, we use the estimated time to compute matrix and the estimated energy consumption matrix which are created using historical data. We design, evaluate, and compare our task scheduling methods to show that a significant improvement in energy consumption can be achieved when considering time-of-use dependent scheduling algorithms. The simulation results show that we improve the performance and the energy values up to 49% when compared to schedulers that do not consider the value functions. Similar to the simulation results, our experimental results from running our value based scheduling on an IBM blade server show up to 82% improvement in performance value, 110% improvement in energy value, and up to 77% improvement in VoS compared to schedulers that do not consider the value functions.

AB - Task scheduling for large-scale computing systems is a challenging problem. From the users perspective, the main concern is the performance of the submitted tasks, whereas, for the cloud service providers, reducing operation cost while providing the required service is critical. Therefore, it is important for task scheduling mechanisms to balance users’ performance requirements and energy efficiency because energy consumption is one of the major operational costs. We present a time dependent value of service (VoS) metric that will be maximized by the scheduling algorithm that take into consideration the arrival time of a task while evaluating the value functions for completing a task at a given time and the tasks energy consumption. We consider the variation in value for completing a task at different times such that the value of energy reduction can change significantly between peak and non-peak periods. To determine the value of a task completion, we use completion time and energy consumption with soft and hard thresholds. We define the VoS for a given workload to be the sum of the values for all tasks that are executed during a given period of time. Our system model is based on virtual machines, where each task will be assigned a resource configuration characterized by the number of the homogeneous cores and amount of memory. For the scheduling of each task submitted to our system, we use the estimated time to compute matrix and the estimated energy consumption matrix which are created using historical data. We design, evaluate, and compare our task scheduling methods to show that a significant improvement in energy consumption can be achieved when considering time-of-use dependent scheduling algorithms. The simulation results show that we improve the performance and the energy values up to 49% when compared to schedulers that do not consider the value functions. Similar to the simulation results, our experimental results from running our value based scheduling on an IBM blade server show up to 82% improvement in performance value, 110% improvement in energy value, and up to 77% improvement in VoS compared to schedulers that do not consider the value functions.

KW - Energy efficient resource allocation

KW - Performance metrics

KW - Resource management

KW - Task scheduling

KW - Value of service

KW - Virtual machines

UR - http://www.scopus.com/inward/record.url?scp=85019596343&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019596343&partnerID=8YFLogxK

U2 - 10.1007/s10586-017-0901-9

DO - 10.1007/s10586-017-0901-9

M3 - Article

AN - SCOPUS:85019596343

SP - 1

EP - 18

JO - Cluster Computing

JF - Cluster Computing

SN - 1386-7857

ER -