A Simulation Framework for Studying Economic Resource Management in Grids

2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (Ccgrid 2012)

2012

Self Cite

A discrete-event simulator's ability to distribute the execution of a simulation model allows one to deal with the memory limitations of a single computational resource, and thereby increase the scale or level of detail at which models can be studied. In addition, distribution has the potential to reduce the round trip time of a simulation by incorporating multiple computational cores into the simulation's execution. However, such gains can be voided by the overhead that time synchronization protocols introduce. These protocols are required to prevent the occurrence of causality errors during a parallel execution of a simulation. The overhead depends on the protocol, characteristics of the simulation model, and the architecture of the computational resources used. Recently, infrastructure-as-a-service offerings in cloud computing have introduced flexibility in acquiring computational resources on a pay-as-you-go basis. At present, it is unclear to what extent these offerings are suited for the distributed execution of discrete-event simulations, and how the characteristics of different resource types impact the runtime performance of distributed simulations. In this paper we investigate this issue, and assess the performance of different conservative time synchronization protocols on a range of cloud resource types that are currently available on Amazon EC2.

Section: B Deployment Of a Distributed Simulation On Amazon Ec2mentioning

confidence: 99%

Conservative Distributed Discrete Event Simulation on Amazon EC2

Vanmechelen

Munck

2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (Ccgrid 2012)

2012

Self Cite

“…We have evaluated the proposed scheduling algorithms in a simulated market environment delivered by the Grid Economics Simulator (GES) [43]. GES is a Java-based simulator that has been developed in order to support research into different market organizations for economic cluster and grid resource management.…”

Section: Grid Economics Simulatormentioning

confidence: 99%

An evaluation of the benefits of fine-grained value-based scheduling on general purpose clusters

Bossche

Vanmechelen

Future Generation Computer Systems

2011

Self Cite

a b s t r a c tGeneral purpose compute clusters are used by a wide range of organizations to deliver the necessary computational power for their processes. In order to manage the shared use of such clusters, scheduling policies are installed to determine if and when the jobs submitted to the cluster are executed. Valuebased scheduling policies differ from other policies in that they allow users to communicate the value of their computation to the scheduling mechanism. The design of market mechanisms whereby users are able to bid for resources in a fine-grained manner has proven to be an attractive means to implement such policies. In the clearing phase of the mechanism, supply and demand for resources are matched in pursuit of a value-maximizing job schedule and resource prices are dynamically adjusted to the level of excess demand in the system. Despite their success in simulations and research literature, such finegrained value-based scheduling policies have been rarely used in practice as they are often considered too fragile, too onerous for end-users to work with, and difficult to implement. A coarse-grained form of value-based scheduling that mitigates the aforementioned disadvantages involves the installation of a priority queuing system with fixed costs per queue. At present, however, it is unclear whether such a coarse-grained policy underperforms in value realization when compared to fine-grained scheduling through auctions, and if so, to what extent. Using workload traces of general purpose clusters we make the comparison and investigate under which conditions efficiency can be gained with the fine-grained policy.

“…The market equilibrium point is the zero of this surface and sets the price at which the market will trade at that moment in time. The algorithm for computing this equilibrium is based on Smale's method [9], [10], with a number of modifications and extensions introduced in [8], [4].…”

Section: E Pricing Algorithmmentioning

confidence: 99%

“…It was found to work well in a wide variety of workloads and infrastructural arrangements. We will use the Grid Economics Simulator (GES) developed in [4] and extended in [8] to run similar simulations. With a number of techniques, support vector machines (SVM), back propagation neural networks (NNBP), sequence pattern (SP), we will model the price evolution an evaluate the quality of the models and of their predictive capabilities.…”

Section: Introductionmentioning

confidence: 99%

Modeling Resource Prices in Grid Markets

Arfa

Hellinckx

2012 Seventh International Conference on P2P, Parallel, Grid, Cloud and Internet Computing

2012

Economic principles have been proposed in the literature to underpin the algorithms for managing resources in complex distributed systems such as grids or peer-to-peer systems. Such principles represents a shift away from traditional system-centric management towards user-centric management. It allows each user to formulate an individual valuation of the requested resources by indicating an acceptable price level for their usage.In a computational commodity market, the price for CPU usage is set by equilibrating supply and demand. We investigate the possibility of modeling the price evolution in a computational market with number of techniques (support vector machines, back propagation neural networks and sequence patterns). If such models have predictive capability, they may be used to optimize resource usage.This study extends previous work. It now includes more differentiated resource model with several categories of CPU's differing in their performance levels and it also adds new modeling approaches for the CPU price evolution. Our study indicates no significant difference in the capabilities of the approaches.