Task assignment for distributed computing

Scholz, Peter; Harbeck, E.

doi:10.1109/apdc.1997.574043

Cited by 4 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Static task scheduling algorithms that satisfy these assumptions fall into three categories: optimal approaches [22], guided random search based algorithms such as genetic algorithms and simulated annealing [13,23,25,29], and heuristics. In our system, task placement is run repeatedly over a dynamic task graph, so efficient run times are essential.…”

Section: Related Workmentioning

confidence: 99%

Incremental placement of interactive perception applications

Yigitbasi

Mummert

Pillai

et al. 2011

Proceedings of the 20th International Symposium on High Performance Distributed Computing

View full text Add to dashboard Cite

Interactive perception applications, such as gesture recognition and vision-based user interfaces, process high-data rate streams with compute intensive computer vision and machine learning algorithms. These applications can be represented as data flow graphs comprising several processing stages. Such applications require low latency to be interactive so that the results are immediately available to the user. To achieve low latency, we exploit the inherent coarsegrained task and data parallelism of these applications by running them on clusters of machines. This paper addresses an important problem that arises: how to place the stages of these applications on machines to minimize the latency, and in particular, how to adjust an existing schedule in response to changes in the operating conditions (perturbations) while minimizing the disruption in the existing placement (churn). To this end, we propose four incremental placement heuristics which use the HEFT scheduling algorithm as their primary building block. Through simulations and experiments on a real implementation, using diverse workloads and a range of perturbation scenarios, we demonstrate that dynamic adjustment of the schedule can improve latency by as much as 36%, while producing little churn.

show abstract

Section: Related Workmentioning

confidence: 99%

Incremental placement of interactive perception applications

Yigitbasi

Mummert

Pillai

et al. 2011

Proceedings of the 20th International Symposium on High Performance Distributed Computing

View full text Add to dashboard Cite

show abstract

“…With this work, we have applied our learning from SARS as a complex adaptive system to design algorithms for task allocation in a distributed system. Other researchers have also taken an inspiration from nature to solve the problem of task allocation 8,9 , but this work highly focuses on reduction of the overall cost of task allocation by understanding the inter-system behaviour and non-linearity of distributed system rather than just following conventional sequential approach.…”

Section: Iso Computers Slips To Isolation If Value = 1 Killmentioning

confidence: 99%

Epidemiology-based Task Assignment Algorithm for Distributed Systems

Brahmbhatt

Camorlinga

2016

Procedia Computer Science

View full text Add to dashboard Cite

Objective: Design task assignment algorithms based on the patterns of disease spread among the population. Scope: Epidemiology studies spatiotemporal patterns of illness in populations and the factors affecting it. An epidemic emerges out of the population activities and environment. Task assignment is a common activity in many realms where sub-tasks are created, delegated, and collectively carried out to achieve the original task. Due to its complexity and context, task assignment can be a challenging activity that can result in limited outcomes. This research studies task assignment as an epidemic assigned to a distributed system. We have developed computational models to understand the outbreak of aerosol-borne diseases by using the agent-based modelling approach. Experiments are carried out to observe the patterns of emergence during the spread of disease among the individuals and get insights of their mechanisms. These mechanisms are used to design algorithms for task assignment on distributed systems. Results: Understanding the emergent behaviour of diseases can provide the platform for the development of distributed algorithms that can be helpful in overcoming some of the challenges of task assignment in a distributed system.

show abstract