Mahantesh N. Birje

Reliable resources brokering scheme in wireless grids based on non-cooperative bargaining game

Wireless computational grids are evolving due to rapid growth in wireless devices to provide any time any where computing and communication services. Wireless devices are considered as important resources for job execution in a wireless grid. However, they are unreliable due to unstable wireless connections, frequent mobility, and device resource constraints such as limited processing capacity, limited battery life, low communication bandwidth, etc. Consumers and providers in a wireless grid have potential conflicting interests: consumers prefer reliable resources at minimum cost for job execution whereas providers prefer efficient utilization of resources with a maximum profit. Hence it is necessary to design a resource brokering mechanism that assures job(s) execution and satisfies needs of consumers and providers. This paper proposes a software agent based resource brokering scheme that employs a resource brokering agency consisting of a set of software agents which discover and recommend reliable devices, negotiate resource cost using non-cooperative bargaining game, and schedule jobs. Device reliability is modeled by using parameters like processing rate, memory, bandwidth and battery power. Discount function is employed in bargaining game based on job deadline, previous offers of an opponent, and dynamics in the grid market. Adaptive negotiation process is applied by exercising different discount strategies (aggressive or conservative or linear) based on the negotiation status of a player. The scheme is simulated to evaluate performance parameters like resource utilization rate, expected profit, job completion time and job execution rate. We observed that the proposed scheme performs better than the existing resource brokering scheme.

Agent based Resource Brokering and Allocation in Wireless Grids

This paper presents an agent based model for discovery, brokering and allocation of cost effective resources to computational jobs. A wireless grid environment of virtual organizations is considered. Agents are employed to perform resource brokering and allocation tasks. The model is simulated for different wireless grid scenarios, to test its operational effectiveness

Wireless Grid Computing: A Survey

Abstract Grid computing enables transparent access to shared and/or idle computing, storage, and network resources anywhere, anytime, to any grid user with guaranteed Quality of Service (QoS). Wireless grids extend the capability of grid computing to wireless devices. The number of users using laptops, PDAs, cell phones, and other wireless devices is increasing leading to more networked wireless devices, and creating a vast collective potential of unexploited resources. Wireless grid computing with its model of coordinated resource sharing may provide a way to utilize such resources that are normally distributed throughout a grid. Grid Computing will be the major area of focus in the future days. We may have Gridnet in the future as we have Internet today. In this survey paper, we tried to cover the entire spectrum of wireless grids: wireless grid computing, its architectures, challenges, communication paradigms, applications, and different standards related to grids.

Monitoring and status representation of devices in wireless grids

Grid Computing is a concept, a network, a work in progress, part hype and part reality, and it is increasingly capturing the attention of the computing community The advancements in wireless technologies and increased number of wireless device users supported the evolution of wireless grids Grid information server (GIS) has to maintain the most up-to-date resource status information of all devices, so that, application can be scheduled to devices that meet its resource requirements. Each wireless device is resource constrained, and its resource status keeps on varying dynamically depending upon number of applications it is executing, amount of data it is communicating, battery level, and mobility In order to keep up-to-date resource status, a continuous monitoring is needed The increase in number of status delivery of such monitored observations will consume lot much of bandwidth, making the database size of grid information server to grow continuously over a period of time. To solve this problem, we consider moderate number of communications of status updates that balances both bandwidth consumption and resource status accuracy Also, we propose three methods to represent these update messages so that bandwidth requirement and latency of communication with GIS is reduced Normal representation, Variable bit length representation, and Relative difference representation methods are proposed and analyzed Relative difference method is analyzed in best case as well as in worst case, and is found to be more efficient compared to other two methods in terms of memory requirements.

WiGriMMA: A Wireless Grid Monitoring Model Using Agents

The complexity, heterogeneity, device mobility and the unpredictable user behavior demands proper automation of monitoring activity in the wireless Grid to enable the user needs. Since the wireless devices can dynamically join/leave the Grid, and its state may be affected by various parameters (like the battery power, signal strength, the number of jobs submitted to it, device mobility, etc.) leading to overload state, it is essential to monitor the devices so that long term resource planning can be achieved. This paper proposes a Wireless Grid Monitoring Model using Agents (WiGriMMA) that monitor the device mobility and state, communicates the state to Grid information server (GIS), provides the resource availability information, controls the selfish users and the device state so that the device is not overloaded. The model is simulated to test its operation effectiveness considering the performance parameters such as resource availability, resource stability, device state, job execution rate, user behavior and agent overhead. The results show that the proposed WiGriMMA performs better than the existing Grid monitoring model (GridView) in terms of the resource availability, device states and the job execution rate.

Economical job scheduling in wireless grid

Wireless grid can be seen as a market place where consumers would like to execute their jobs and resource providers would like to provide the resources to consumers based on some cost. The status of resources (like CPU, memory) is affected by the factors such as the number of jobs submitted to it, the number of jobs executing, battery power etc. This paper proposes a cost effective job scheduling mechanism based on system state. It considers the dynamically changing resource status, predicts the resource state, estimates the job cost and schedules the job to optimal resources. We simulated different test cases using load traces sampled from Pentium machine during particular time. Our simulation results demonstrate that the proposed job scheduling method is cost effective and reduces the job rejection ratio.

Resource pricing strategy in wireless grid using non-cooperative bargaining game

The consumers in a wireless grid will prefer reliable and cost optimal resources for job execution, while the grid service providers prefer efficient utilization of their resources. Hence brokering of resources that meet the requirements of consumers and providers is a challenging task in presence of unstable (wireless) network connections, market dynamics and rational users. This paper models a resource pricing strategy using a non-cooperative bargaining game for resource allocation considering dynamics in the grid market. The proposed scheme is simulated to evaluate the performance parameters like offered price, surplus, negotiation time and job completion time. We observed that our scheme performs better than existing scheme in terms of negotiation time and job completion time.

An efficient method of sharing device resource status in wireless grids

The resource status of a device in a wireless grid is affected by several factors such as the number of applications in execution, amount of data communication, device mobility, battery power, signal strength, etc. The grid scheduler schedules user applications based on current resource status of a device stored in the grid information server GIS. Thus, it is very much essential to maintain the correct updated status information at GIS for proper scheduling. This paper proposes an efficient method for sharing of resource status with GIS using software agents to facilitate the execution of compute intensive tasks that depend more on processor than on memory or bandwidth. At each device, it employs a device agency with three agents called as Status Monitoring Agent, Status Representation Agent, and Status Communication Agent. These agents monitor the changes in resource status, store status information in compact manner, and share the changed status with GIS efficiently so that status accuracy is achieved as well as the bandwidth and memory consumption is reduced. The proposed work is simulated to evaluate the performance parameters such as memory and bandwidth requirements, memory reduction rate and redundancy. It is observed that the proposed work achieved resource status accuracy by sharing relevant status changes in timely manner and reduced bandwidth and memory requirements.

Resource discovery and allocation in wireless grids by using a multiagent system

This paper presents an Agent-based Discovery and Allocation of Resources (ADAR) model for wireless grids of virtual organizations. ADAR allocates cost effective resources for the jobs in order to maximize the resource utilization. It employs resource discovery mechanism in a hierarchical fashion: firstly within a local cluster, secondly within a virtual organization if a resource is not available in the local cluster, and finally among the virtual organizations if a resource is not available within the local virtual organization. ADAR uses five types of agents namely Job Processing Agents (JPAs), Job Mobile Agents (JMAs), Resource Monitoring Agents (RMAs), Actual Organization Resource Broker Agents (AORBAs), and Virtual Organization Resource Manager Agents (VORMAs). JPA is static and executes a job within the same machine if the resources are available; otherwise it creates a JMA that carries the job requirements with it and communicates with its associated AORBA to discover the required resources. The AORBA is a static agent and discovers the optimum cost resources within its cluster. AORBA interacts with the VORMAs in case the resources are not available within a cluster. VORMA is a static agent and can interact with other VORMAs. The RMA is a static agent and monitors the resource status of a machine. The discovery results are informed to JMA, which makes the source device to migrate its job to the discovered resource. ADAR is evaluated in a simulated environment by using different wireless grid scenarios for virtual organizations. The performance parameters evaluated are: job cost function, resource utilization, bandwidth utilization, agent overheads, and resource discovery time against different mobility factors and varying system loads.