Discouraging free riding in a peer-to-peer cpu-sharing grid

Grid and voluntary computing are well-established solutions for delivering high-performance computational power to e-Science projects. However, not all labs can effectively benefit from these technologies, as they would need to have high-visibility projects and/or highly qualified personnel. The OurGrid proposal is to offer a complete solution for building an open free-to-join grid, where no human interaction is required to allow one to join the grid. In this paper we report our experience in designing, implementing and deploying a grid infrastructure based on the OurGrid middleware. In particular, we focus on two main challenges: security and scalability. We discuss these issues and present a brief description of the solutions proposed by us to address them.

akamaiuniversity.us

Currently, both Peer-to-Peer Computing (P2P) and Grid Computing have remained the most vibrant and useful forms of distributed computing all over the world. Their applications are such that they cut across both academia and industry. It has come to the notice of researchers that there are great misunderstanding and misinterpretation on what these forms of distributed computing actually portend and stand for. In this paper therefore, we take a critical look at comparative study of both computing technologies with aim of making readers understand in a clear cut what each really stands for. To have a good comparison, we start by giving a well referenced definition of Grid Computing as well as Peer-to-Peer Computing. Also, we used technical issues and general features in our comparison vis-à-vis the architecture, security issue, data movement, application deployment, and operating system requirement. We also considered the strength of both distributed computing system and finally we considered what could be the future of both technologies.

2010 39th International Conference on Parallel Processing Workshops, 2010

In this paper, we propose a new algorithm for job interchange in Computational Grids that consist of autonomous and equitable HPC sites, called Shaking-G. Originally developed for balancing the sharing of video files in P2P networks, we conceptually transfer and adapt the algorithm to the domain of job scheduling in Grids, building an integrated, load-adaptive two-tier job exchange strategy. We evaluate the performance of Shaking-G with real workload data in different experimental scenarios and show that it outperforms traditional non-Grid aware algorithms in setups without job interchange, fostering the benefits of collaboration between large HPC centers.

Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID'06), 2006

Typically, users of the Grid (i.e. resource consumers) identify resources to submit their tasks based on the resources' capabilities, e.g. processor type, memory size, etc. However, the reliability of the users who manage the resources (i.e. resource providers) is usually unknown to the consumers, e.g. one provider may intentionally drop a consumer's submitted task in favor of another consumer's task since the latter can pay more. Therefore, there is a risk for consumers that their submitted tasks may fail. Establishing reputation systems is a good alternative to manage this risk [1]. However, existing work regarding the reputation systems in the Grid mainly focuses on methods to calculate reputation to model resource providers' reliability. They do not effectively mitigate the influence of inaccurate testimonies and malicious referrers, which are common phenomena in real environments. Against this background, this paper proposes a robust reputation system that has components to mitigate the influence of the inaccurate testimonies and malicious referrers. Experimental results show that the proposed system is efficient in mitigating the influence of these inaccurate testimonies and malicious referrers.

First EELA-2 Conference, EELA-2, Colombia, 2009

Grid technologies are well-established solutions for delivering high-performance computational power to e-Science projects. However, not all users can effectively benefit from these technologies, as they would need to have high-visibility projects and/or highly qualified personnel. The OurGrid proposal is to offer a complete solution for building an open free-tojoin grid, where no human interaction is required to allow one to join the grid. In this paper we report our experience in designing and implementing the OurGrid middleware. In particular, we focus on the security aspects of OurGrid, since this is one of the main challenges in setting up a peer-to-peer grid. OurGrid is in production since December 2004.

Making Grids Work, 2008

In this paper, we discuss how Peer-to-Peer data distribution techniques can be adapted to Desktop Grid computing environments, particularly to the BOINC platform. To date, Desktop Grid systems have focused primarily on utilizing spare CPU cycles, yet have neglected to take advantage of client network capabilities. Leveraging client bandwidth will not only benefit current projects by lowering their overheads but will also facilitate Destkop Grid adoption by data-heavy applications. In this paper, weoutline two approaches to Peer-to-Peer data sharing that could be adapted for volunteer computing platforms: the highly successful BitTorrent protocol; and, a more complex, yet secure and customizable, hierarchal Peer-to-Peer data center approach.

2008 IEEE International Symposium on Parallel and Distributed Processing, 2008

Research projects in many fields are increasingly reliant on the use of computer-based simulation and computing grids. Many projects have successfully leveraged voluntary computing infrastructures by developing and distributing "@home" applications using the BOINC framework. Through generous contributions from the general public, these systems now have a computing backbone on which to have their data processed or simulations run. A shortcoming of such systems is that most users are often limited to contributing resources and few users are capable of developing or porting their own applications in order to use these resources. While many users are satisfied with receiving points (an intangible good) in return for their contribution, the need to port applications presents a barrier to entry to many other users who can potentially benefit from using the voluntary resources.

Parallel Computing, 2007

This paper presents a Peer-to-Peer (P2P) infrastructure that supports a large scale grid. The P2P infrastructure is implemented in Java and federates Java Virtual Machines (JVMs) for computation. The management of shared JVMs is decentralized, self-organized, and configurable.

2010

Grids facilitate creation of wide-area collaborative environment for sharing computing or storage resources and various applications. Interconnecting distributed Grid sites through peer-to-peer routing and information dissemination structure (also known as Peer-to-Peer Grids) is essential to avoid the problems of scheduling efficiency bottleneck and single point of failure in the centralized or hierarchical scheduling approaches. On the other hand, uncertainty and unreliability are facts in distributed infrastructures such as Peer-to-Peer Grids, which are triggered by multiple factors including scale, dynamism, failures, and incomplete global knowledge. In this paper, a reputation-based Grid workflow scheduling technique is proposed to counter the effect of inherent unreliability and temporal characteristics of computing resources in large scale, decentralized Peer-to-Peer Grid environments. The proposed approach builds upon structured peer-to-peer indexing and networking techniques to create a scalable wide-area overlay of Grid sites for supporting dependable scheduling of applications. The scheduling algorithm considers reliability of a Grid resource as a statistical property, which is globally computed in the decentralized Grid overlay based on dynamic feedbacks or reputation scores assigned by individual service consumers mediated via Grid resource brokers. The proposed algorithm dynamically adapts to changing resource conditions and offers significant performance gains as compared to traditional approaches in the event of unsuccessful job execution or resource failure. The results evaluated through an extensive trace driven simulation show that our scheduling technique can reduce the makespan up to 50% and successfully isolate the failure-prone resources from the system.

Grid and Cooperative Computing, 2003

The common goals of the Grid and peer-to-peer communi- ties have brought them in close proximity. Both the technologies overlay a collaborative resource-sharing infrastructure on existing (public) net- works. In realizing this shared goal, however, they concentrate on signif- icantly contrasting issues. The Grid paradigm focuses on performance, control, security, specialization, and standardization. On the other hand, the peer-to-peer paradigm