An Opportunity Cost Approach for Job Assignment in a Scalable Computing Cluster

@article{Amir2000AnOC,
  title={An Opportunity Cost Approach for Job Assignment in a Scalable Computing Cluster},
  author={Yair Amir and Baruch Awerbuch and Amnon Barak and Ryan S. Borgstrom and Arie Keren},
  journal={IEEE Trans. Parallel Distributed Syst.},
  year={2000},
  volume={11},
  pages={760-768}
}
A new method is presented for job assignment to and reassignment between machines in a computing cluster. Our method is based on a theoretical framework that has been experimentally tested and shown to be useful in practice. This "opportunity cost" method converts the usage of several heterogeneous resources in a machine to a single homogeneous "cost." Assignment and reassignment are then performed based on that cost. This is in contrast to traditional, ad hoc methods for job assignment and… Expand
Opportunity Cost Algorithms for Reduction of I/O and Interprocess Communication Overhead in a Computing Cluster
TLDR
These algorithms are designed to improve the overall CPU utilization of the cluster and to reduce the I/O and the interprocess communication (IPC) overhead and outperform existing methods for process allocation that are based on ad hoc heuristics. Expand
Dynamic job Scheduling in Cloud Computing based on horizontal load balancing
Cloud computing is a latest new computing paradigm where applications, data and IT services are provided across dynamic and geographically dispersed organization. How to improve the global throughputExpand
A cost-benefit framework for online management of a metacomputing system
TLDR
The Cost-Benefit Framework is presented, inspired by economic principles, which simplifies complex assignment and admission control decisions, and performs well in practice in an Internet-wide market for computational services. Expand
A Cost-Benefit framework for online management of a metacomputing system
TLDR
This framework simplifies complex assignment and admission control decisions, and performs well in practice in the context of an Internet-wide market for computational services and for a classic network of workstations. Expand
A Cost-Based Online Scheduling Algorithm for Job Assignment on Computational Grids
TLDR
This paper expands on the previous work, which considers CPU and memory with economic principle in the cluster environment, and presents a cost-based online scheduling algorithm for job assignment in the grid environment. Expand
Heuristics for Resource Matching in Intel's Compute Farm
TLDR
Investigation of resource matching between jobs and machines in Intel's compute farm shows that common heuristics such as Best-Fit and Worse-Fit may fail to properly utilize the available resources when applied to either cores or memory in isolation, and proposes Mix-Fit, a heuristic which attempts to balance usage between resources. Expand
The home model and competitive algorithms for load balancing in a computing cluster
  • R. Lavi, A. Barak
  • Computer Science
  • Proceedings 21st International Conference on Distributed Computing Systems
  • 2001
TLDR
This work defines the home model in order to better reflect the architecture of a computing cluster and develops several online algorithms for load balancing in this model, which is the best existing theoretical model to describe such clusters. Expand
A stochastic load balancing algorithm for i‐Computing
This paper presents a stochastic dynamic load balancing algorithm for Internet computing, which is a new type of distributed computing involving heterogeneous workstations from differentExpand
Task-Scheduling in Cloud Computing using Credit Based Assignment Problem
TLDR
Credit based scheduling decision is used to evaluate the entire group of task in the task queue and find the minimal completion time of all task and cost matrix has been generated as the fair tendency of a task to be assigned in a resource. Expand
Survey and analysis of optimal scheduling strategies in cloud environment
  • M. Paul, G. Sanyal
  • Computer Science
  • 2011 World Congress on Information and Communication Technologies
  • 2011
TLDR
This paper has proposed a methodology by using assignment to assign these jobs to the suitable resources so as to minimize the whole completion time. Expand
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 18 REFERENCES
On-line load balancing with applications to machine scheduling and virtual circuit routing
TLDR
An algorithm is described that achieves an O (log n) competitive ratio, where n is the number of nodes in the network, for the case where virtual circuits continue to exist forever and for the related machines case, the first algorithm that achieves constant competitive ratio is described. Expand
Competitive routing of virtual circuits with unknown duration
TLDR
It is shown that by allowing O(logn) reroutes per call, the routing algorithm can achieve O( logn) competitive ratio with respect to the maximum load for the unknown duration case, where n is the number of nodes in the network. Expand
The MOSIX multicomputer operating system for high performance cluster computing
TLDR
The performance of the MOSIX operating system with algorithms for adaptive resource sharing as well as the performance of several large-scale, parallel applications are presented. Expand
Exploiting process lifetime distributions for dynamic load balancing
TLDR
It is found that, contrary to previous reports, the performance benefits of preemptive migration are significantly greater than those of non-preemptive migration, even when the memory-transfer cost is high. Expand
Exploiting process lifetime distributions for dynamic load balancing
TLDR
It is found that, contrary to previous reports, the performance benefits of preemptive migration are significantly greater than those of non-preemptive migration, even when the memory-transfer cost is high. Expand
Packet Routing via Min-Cost Circuit Routing
TLDR
This paper introduces competitive algorithms for a new problem called min-cost load circuit routing, here, the goal is to create on-line virtual circuits in a graph, while trying to simultaneously minimize link bandwidth and (related) communication costs. Expand
On-line routing of virtual circuits with applications to load balancing and machine scheduling
TLDR
An algorithm is described that achieves on-line allocation of routes to virtual circuits (both point-to-point and multicast) with a constant competitive ratio with respect to maximum congestin, where n is the number of nodes in the network. Expand
New algorithms for an ancient scheduling problem
TLDR
To the best of the knowledge, this 4/3-competitive algorithm is the first specifically randomized algorithm for the original, original, on-line scheduling problem, and probably the first directly randomized algorithm of its kind. Expand
aOn-Line Routing of Virtual Circuits with Applications to Load Balancing and Machine Scheduling,o
  • J. ACM,
  • 1997
On-Line Machine Scheduling with Applications to Load Balancing and Virtual Circuit Routing
  • Proceedings of the ACM Symposium on Theory Of Computing (STOC),
  • 1993
...
1
2
...