Opportunity cost algorithms for reduction of I/O and interprocess communication overhead in a computing cluster

Keren, Arie; Barak, Amnon

doi:10.1109/tpds.2003.1167369

Cited by 32 publications

(15 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Process reassignments can be triggered by a change in the number of processes or their profiles, or a change in the utilization or number of cluster nodes. An opportunity cost algorithm is used to decide whether processes should be migrated and, if so, to which cluster nodes [59].…”

Section: Mosix and Derivativesmentioning

confidence: 99%

Single system image: A survey

Healy

Lynn

Barrett

et al. 2016

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Section: Mosix and Derivativesmentioning

confidence: 99%

Single system image: A survey

Healy

Lynn

Barrett

et al. 2016

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

“…To overcome this problem of task allocation in a distributed environment more than single task will assign to a processor in order to get execute minimum response time. Some of the task allocation methods have been reported in the literature, such as Such as processing reliability [8,10], Task Allocation for Minimizing Cost [1,2,4,5,6,7,11,12,13,15,16,17], Evaluation of gang scheduling performance and cost [9], Decentralized management of bi-modal network resource in a distributed stream processing platform [3] and Static workload distribution of parallel application [14]. In this research paper we have taken the problem of task allocation in distributed computing environment based on execution time.…”

Section: Figure 1: Task Allocation Problemmentioning

confidence: 99%

Static Approach for Efficient Task Allocation in Distributed Environment

Khan¹,

Govil²

2013

IJCA

View full text Add to dashboard Cite

Task allocation in a distributed environment is a fundamental problem. In such environment an application runs on a single computer can be accessible on every system / terminal present in the same network. Therefore geographical distance does not matter. If number of tasks is less than or equal to available processors in the network, we can assign these task without any issue. But this allocation becomes complex when numbers of tasks are greater than the numbers of processors. The problem of task allocation for processing of 'm' tasks to 'n' processors (m>n) in a distributed networks is addressed here through a new modified tasks allocation technique. The algorithm, presented in this paper allocates the tasks to the processor of different processing capacity to increase the performance of the distributed network. The Algorithm addressed in this paper is based on the consideration of processing time of the task to the processors. Keeping this in view we have suggested a new algorithm to assign all the tasks as per the required availability of processors and their processing capacity so that none of the tasks get remains unexecuted in the distributed environment.

show abstract

“…Communication-aware load balancing has been studied in Cruz and Park [2001] and Keren and Barak [2003]. Our approach takes into account the communication load as a measure to determine the migration cost, but balancing the network load is beyond the scope of this paper.…”

Section: Dynamic Load Balancing For I/o-intensive Applications On Clumentioning

confidence: 99%

Dynamic load balancing for I/O-intensive applications on clusters

et al. 2009

View full text Add to dashboard Cite

Load balancing for clusters has been investigated extensively, mainly focusing on the effective usage of global CPU and memory resources. However, previous CPU-or memory-centric load balancing schemes suffer significant performance drop under I/O-intensive workloads due to the imbalance of I/O load. To solve this problem, we propose two simple yet effective I/O-aware load-balancing schemes for two types of clusters: (1) homogeneous clusters where nodes are identical and (2) heterogeneous clusters, which are comprised of a variety of nodes with different performance characteristics in computing power, memory capacity, and disk speed. In addition to assigning I/O-intensive sequential and parallel jobs to nodes with light I/O loads, the proposed schemes judiciously take into account both CPU and memory load sharing in the system. Therefore, our schemes are able to maintain high performance for a wide spectrum of workloads. We develop analytic models to study mean slowdowns, task arrival, and transfer processes in system levels. Using a set of real I/O-intensive parallel applications and synthetic parallel jobs with various I/O characteristics, we show that our proposed schemes consistently improve the performance over existing non-I/O-aware load-balancing schemes, including CPU-and Memory-aware schemes and a PBS-like batch scheduler for parallel and sequential jobs, for a diverse set of workload conditions. Importantly, this performance improvement becomes much more pronounced when the applications are I/O-intensive. For example, the proposed approaches deliver 23.6-88.0 % performance improvements for I/Ointensive applications such as LU decomposition, Sparse Cholesky, Titan, Parallel text searching, and Data Mining. When I/O load is low or well balanced, the proposed schemes are capable of maintaining the same level of performance as the existing non-I/O-aware schemes.

show abstract

Opportunity cost algorithms for reduction of I/O and interprocess communication overhead in a computing cluster

Cited by 32 publications

References 37 publications

Single system image: A survey

Single system image: A survey

Static Approach for Efficient Task Allocation in Distributed Environment

Dynamic load balancing for I/O-intensive applications on clusters

Contact Info

Product

Resources

About