Population-Based Metaheuristics for Tasks Scheduling in Heterogeneous Distributed Systems

Zamfirache, Flavia; Frîncu, Marc; Zaharie, Daniela

doi:10.1007/978-3-642-18466-6_38

Cited by 5 publications

(6 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Algunos de los métodos de búsqueda local para la planificación, basados en el enfoque de la vecindad son: movimiento de la base con búsqueda local [48] , intercambio de base basado en búsqueda local [49] y porúltimo se cita al método de Búsqueda Variable del Vecindario -Variable Neighborhood Search (VNS) [50].…”

Section: A Enfoque De Heurística Basada En Búsqueda Localunclassified

Estrategias de planificación para datos y procesos en computación Grid: estado del arte

Bermúdez

Pérez

2019

Ing.USBMed

View full text Add to dashboard Cite

Resumen. La distribución y la naturaleza compartida y heterogénea de la computación grid, hace que su objetivo de ofrecer aplicaciones con poder computacional colectivo, sea un gran reto. El objetivo es presentar el resultado de una investigación sistemática sobre aspectos teóricos de las temáticas que convergen en el diseño y desarrollo de planificadores de datos y procesos en computación grid. Se tomó como referente la metodología planteada para desarrollo de estados de arte, que cuenta con dos fases principales: heurística y hermenéutica. En el proceso se analizaron los fundamentos teóricos, tales como: la arquitectura, las etapas que conforman el proceso realizado por parte de un planificador grid y los tipos de planificación existentes, todo ello con el objeto de abordar los diferentes modelos y enfoques computacionales, heurísticos y metaheurísticos que permiten un funcionamiento eficiente y eficaz de maneraóptima de los planificadores grid, permitiendo mitigar en cierto grado los problemas presentados en la planificación grid. La optimización de la planificación de recursos grid, es unárea que se encuentra en desarrollo, dado que las problemáticas principales como: demora en la planificación y el proceso de optimización, están aún en etapa de desarrollo. Palabras Clave. Computación grid; heurística; metaheurística; planificación grid; procesos; trabajos.Abstract. The distribution, the shared and heterogeneous nature of grid computing, makes its objective of offering applications with a collective computational power, be a great challenge. The objetive is to present the result of the systematic research about theoretical aspects about converge topics in the design and development of data and process planners in GRID computation. The methodology proposed for the development of art states was taken as a reference, which has two main phases: heuristics and hermeneutics. In this process, the theoretical foundations are analyzed like the architecture, the stages that conforms the process performed by a GRID planner and the existing planner types; all this with the aim of addressing the different computational, heuristic and metaheuristic models and approaches that allows the efficient and effective operation of GRID planners, allowing to mitigate in certain degree the problems presented in GRID planning. The optimization of grid resource planning is an area that is under development, given that the main problems such as: delay in planning and the optimization process, are still in the development stage.

show abstract

Section: A Enfoque De Heurística Basada En Búsqueda Localunclassified

Estrategias de planificación para datos y procesos en computación Grid: estado del arte

Bermúdez

Pérez

2019

Ing.USBMed

View full text Add to dashboard Cite

show abstract

“…Other approximate procedures such as gradient descent are not directly applicable to the problem at hand since the optimization problem presents a non-differentiable, non-continuous function. However, nature-inspired search algorithms such as Genetic Algorithms (GAs), ant-colony optimization, and particleswarm optimization, which follow a guided randomized search, are sometimes more effective than traditional heuristics in solving similar single and multi-objective workflow scheduling problems [24], [25], [26], [27]. In this article, we also explore GAs for finding an approximate solution to the optimization problem.…”

Section: Problem Formulationmentioning

confidence: 99%

“…Meta-heuristics have been widely used to address the task scheduling problem [24], [25], [26], [27]. Most of the approaches are nature inspired and rely on GA [28], ant colony optimization [49], particle swarm optimization [50] or simulated annealing [51] techniques to search for sub-optimal solutions.…”

Section: Related Workmentioning

confidence: 99%

Reactive Resource Provisioning Heuristics for Dynamic Dataflows on Cloud Infrastructure

Kumbhare

Simmhan

Frîncu

et al. 2015

IEEE Trans. Cloud Comput.

View full text Add to dashboard Cite

The need for low latency analysis over high-velocity data streams motivates the need for distributed continuous dataflow systems. Contemporary stream processing systems use simple techniques to scale on elastic cloud resources to handle variable data rates. However, application QoS is also impacted by variability in resource performance exhibited by clouds and hence necessitates autonomic methods of provisioning elastic resources to support such applications on cloud infrastructure. We develop the concept of "dynamic dataflows" which utilize alternate tasks as additional control over the dataflow's cost and QoS. Further, we formalize an optimization problem to represent deployment and runtime resource provisioning that allows us to balance the application's QoS, value, and the resource cost. We propose two greedy heuristics, centralized and sharded, based on the variable-sized bin packing algorithm and compare against a Genetic Algorithm (GA) based heuristic that gives a near-optimal solution. A large-scale simulation study, using the Linear Road Benchmark and VM performance traces from the AWS public cloud, shows that while GA-based heuristic provides a better quality schedule, the greedy heuristics are more practical, and can intelligently utilize cloud elasticity to mitigate the effect of variability, both in input data rates and cloud resource performance, to meet the QoS of fast data applications.

show abstract

“…Genetic Algorithms (GA) proved to give good results [37,38] when scheduling Grid applications. It comes with no surprise that they have been widely proposed for Cloud systems also.…”

Section: Related Work On High Available Systems and Cloud Schedulingmentioning

confidence: 99%

Scheduling highly available applications on cloud environments

Frîncu

2014

Future Generation Computer Systems

View full text Add to dashboard Cite

Cloud computing is becoming a popular solution for storing data and executing applications due to its on-demand pay-per-use policy that allows access to virtually unlimited resources. In this frame applications such as those oriented towards Web 2.0 begin to be migrated on cloud systems. Web 2.0 applications are usually composed of several components that run indefinitely and need to be available to end users throughout their execution life cycle. Their availability strongly depends on the number of resource failures and on the variation in user hit rate. These problems are usually solved through scaling. A scaled application can span its components on several nodes. Hence if one or more nodes fail it could become unavailable. Therefore we require a method of ensuring the application's functionality despite the number of node failures. In this paper we propose to build highly available applications, i.e., systems with low downtimes, by taking advantage of the component based architecture and of the application scaling property. We present a solution to finding the optimal number of component types needed on nodes so that every type is present on every allocated node. Furthermore nodes cannot exceed a maximum threshold and the total running cost of the applications needs to be minimized. A sub-optimal solution is also given. Both solutions rely on genetic algorithms to achieve their goals. The efficiency of the suboptimal algorithm is studied with respect to its success rate, i.e., probability of the schedule to provide highly available applications in case all but one node fail. Tests performed on the sub-optimal algorithm in terms of node load, closeness to the optimal solution and success rate prove the algorithm's efficiency.

show abstract

Population-Based Metaheuristics for Tasks Scheduling in Heterogeneous Distributed Systems

Cited by 5 publications

References 6 publications

Estrategias de planificación para datos y procesos en computación Grid: estado del arte

Estrategias de planificación para datos y procesos en computación Grid: estado del arte

Reactive Resource Provisioning Heuristics for Dynamic Dataflows on Cloud Infrastructure

Scheduling highly available applications on cloud environments

Contact Info

Product

Resources

About