Students: Raymond Sweha, Vatche Ishakian
\nPIs: Azer Bestavros, John Byers<\/p>\n
This project develops a Cloud Service for Internet content distribution. Our system assists Seeders (content originators) with the dissemination of content (a file) to a set of nodes using Peer-to-Peer (P2P) concepts so that this dissemination is completed in the minimum time possible. Prior results of ours suggest that minimizing content distribution time may be achieved by adding nodes that are not themselves interested in downloading the content, but rather in assisting other nodes with their download in a prescribed (provably optimal) fashion. We call such nodes Angels. The emerging cloud computing architecture offers the best mechanism to allow such Angels to be created on-the-fly. As seeders request assistance with their file distribution, our service responds by spawning virtual machines that act as Angels. In this work, we describe the design and implementation of our “Angels-on-Demand” cloud service as well as the API for invoking this service.<\/p>\n
Students: Vatche Ishakian, Raymond Sweha, Jorge Londo\u00f1o
\nPIs: Azer Bestavros<\/p>\n
By colocating with other tenants of an Infrastructure as a Service (IaaS) offering, IaaS users could reap significant cost savings by judiciously sharing their use of the fixed-size instances offered by IaaS providers. This work presents the blueprints of a Colocation as a Service (CaaS) framework. CaaS strategic services identify coalitions of selfinterested users that would benefit from colocation on shared instances. CaaS operational services provide the information necessary for, and carry out the reconfigurations mandated by strategic services. CaaS could be incorporated into an IaaS offering by providers; it could be implemented as a valueadded proposition by IaaS resellers; or it could be directly leveraged in a peer-to-peer fashion by IaaS users. To establish the practicality of such offerings, this paper presents XCS \u2013 a prototype implementation of CaaS on top of the Xen hypervisor. XCS makes specific choices with respect to the various elements of the CaaS framework: it implements strategic services based on a game-theoretic formulation of colocation; it features novel concurrent migration heuristics which are shown to be efficient; and it offers monitoring and accounting services at both the hypervisor and VM layers. Extensive experimental results obtained by running PlanetLab trace-driven workloads on the XCS prototype confirm the premise of CaaS \u2013 by demonstrating the efficiency and scalability of XCS, and by quantifying the potential cost savings accrued through the use of XCS.<\/p>\n
Students: Vatche Ishakian
\nPIs: Azer Bestavros, Assaf Kfoury<\/p>\n
Desirable application performance is typically guaranteed through the use of Service Level Agreements (SLAs) that specify \ufb01xed fractions of resource capacities that must be allocated for unencumbered use by the application. The mapping between what constitutes desirable performance and SLAs is not unique: multiple SLA expressions might be functionally equivalent. Having the \ufb02exibility to transform SLAs from one form to another in a manner that is provably safe would enable hosting solutions to achieve signi\ufb01cant ef\ufb01ciencies. This work tries to demonstrate the promise of such an approach by proposing a type-theoretic framework for the representation and safe transformation of SLAs.<\/p>\n
Students: Jorge Londo\u00f1o
\nPIs: Azer Bestavros, Shang-Hua Teng<\/p>\n
Distributed computing infrastructures are ubiquitous: applications running on the web rely on Cloud infrastructures, scientists run resource intensive applications on Grid systems, and network architects develop next generation protocols and services on testbeds such as PlanetLab, Emulab\/Netbed, and GENI. A common challenge facing all these as well as other systems is finding and allocating the resources needed in support of such applications and services. This problem is generally known as the network embedding problem<\/em>\/<\/p>\n Embedding Games<\/em> arise when multiple competing agents engage in solving the network embedding problem, with each agent trying to maximize its own benefit. From a modeling and analysis perspective, important questions that arise in this context include determining whether and under what conditions do equilibria governing agent interactions exist, whether convergence to such equilibria is assured, and the extent of the inefficiencies associated with these equilibria. From a system design perspective, the main challenge is that of designing mechanisms that provide incentives that ensure that the choices of the non-cooperating, selfish agents engaged in an embedding game lead to an efficient allocation of resources from a system-wide perspective.<\/p>\n","protected":false},"excerpt":{"rendered":" Angels in The Cloud An on Demand Peer-Assisted Content Distribution Cloud Service Students: Raymond Sweha, Vatche Ishakian PIs: Azer Bestavros, John Byers This project develops a Cloud Service for Internet content distribution. Our system assists Seeders (content originators) with the dissemination of content (a file) to a set of nodes using Peer-to-Peer (P2P) concepts so […]<\/p>\n","protected":false},"author":3458,"featured_media":0,"parent":908,"menu_order":2,"comment_status":"closed","ping_status":"closed","template":"","meta":[],"_links":{"self":[{"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/pages\/917"}],"collection":[{"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/users\/3458"}],"replies":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/comments?post=917"}],"version-history":[{"count":16,"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/pages\/917\/revisions"}],"predecessor-version":[{"id":1508,"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/pages\/917\/revisions\/1508"}],"up":[{"embeddable":true,"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/pages\/908"}],"wp:attachment":[{"href":"https:\/\/www.bu.edu\/riscs\/wp-json\/wp\/v2\/media?parent=917"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}