Abstract:A look-ahead algorithm is proposed to solve the online multiple workflow scheduling problem with two constraints in heterogeneous system. In this problem, workflows come to the system online when they are released, and each workflow is composed of multiple tasks which can be executed on heterogeneous processors, according to their types. Considered two constrains including the non-preemptive processor and the task order in a workflow, the proposed algorithm utilizes the information contained in the list of the… Show more
“…The performance of the distributed system is dependent on the allocation of users' jobs on computing resources. The problem of effectively managing the utilization of resources is referred to as a resource allocation problem, which is very common in production, agriculture, and computer industries (Comito et al, 2013;Deng et al, 2018;Kaur and Bala, 2019;Xu et al, 2017).…”
Resource allocation in a distributed computing system is the process of allocating the workload across multiple computing resources to optimize the required performance criteria. In this article, a resource allocation problem that arises in a distributed system consisting of multiple heterogeneous servers is addressed. The problem is modeled as a multi-objective problem with two conflicting objectives: (a) to minimize the users’ expected response time and (b) to reduce the utilization imbalance between servers. To satisfy these objectives simultaneously, first, both the objectives are considered in an integrated manner, and an optimization problem is formulated. Second, the optimization problem is cast into a game-theoretic setting and modeled as a non-cooperative game, called a non-cooperative resource allocation game. Finally, to solve the game, a differential evolution-based co-evolutionary framework (DECEF) is proposed. To evaluate the performance of DECEF, a rigorous simulation study is carried out. Furthermore, to assess the relative performance of DECEF, it is compared against two existing approaches, from various aspects, including system utilization, system heterogeneity, and system size. The experimental results show that DECEF provides better system-wide performance while optimizing both the objectives.
“…The performance of the distributed system is dependent on the allocation of users' jobs on computing resources. The problem of effectively managing the utilization of resources is referred to as a resource allocation problem, which is very common in production, agriculture, and computer industries (Comito et al, 2013;Deng et al, 2018;Kaur and Bala, 2019;Xu et al, 2017).…”
Resource allocation in a distributed computing system is the process of allocating the workload across multiple computing resources to optimize the required performance criteria. In this article, a resource allocation problem that arises in a distributed system consisting of multiple heterogeneous servers is addressed. The problem is modeled as a multi-objective problem with two conflicting objectives: (a) to minimize the users’ expected response time and (b) to reduce the utilization imbalance between servers. To satisfy these objectives simultaneously, first, both the objectives are considered in an integrated manner, and an optimization problem is formulated. Second, the optimization problem is cast into a game-theoretic setting and modeled as a non-cooperative game, called a non-cooperative resource allocation game. Finally, to solve the game, a differential evolution-based co-evolutionary framework (DECEF) is proposed. To evaluate the performance of DECEF, a rigorous simulation study is carried out. Furthermore, to assess the relative performance of DECEF, it is compared against two existing approaches, from various aspects, including system utilization, system heterogeneity, and system size. The experimental results show that DECEF provides better system-wide performance while optimizing both the objectives.
“…Each client connection occurs at different time with different resource requirements. The varying interactions exert unique challenges for cloud resource usage (Xu et al, 2017).…”
Cloud computing makes scientists to run complex scientific applications. The research community is able to access on-demand compute resources within a short span of time instead of experiencing peak demand bottlenecks. As the demand for cloud resources is dynamic and volatile in nature, this in turn affects the availability of resources during scheduling. In order to allocate sufficient resources for scientific applications with different execution requirements, it is necessary to predict the appropriate set of resources. To attain this objective, a resource prediction–based scheduling technique has been introduced which automates the resource allocation for scientific application in virtualized cloud environment. First, the proposed prediction model is trained on the dataset generated by concurrently deploying tasks of a scientific application on cloud. Then, the resources are scheduled based on the output of proposed prediction technique. The main objective of resource prediction–based scheduling technique is to efficiently handle the resources for virtual machines in order to reduce the execution time, error rate, and improve the accuracy.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.