A Case for Informed Service Placement on PlanetLab
نویسندگان
چکیده
We examine PlanetLab resource utilization data from the perspective of the designer of a resource discovery system. We are particularly interested in variability of, and correlations among, resource utilization attributes, both among nodes and over time. We find that, for some resources, the quantity available at a fixed time differs significantly across nodes, suggesting a potential benefit to using a resource discovery system to wisely place application instances. We further find that, for many nodes, resource availability varies over time, suggesting periodic migration of application instances could be useful. We find that some attributes are moderately correlated among nodes at the same site, but that on a single node, attributes are generally not correlated to one another. Finally, we find that for some node pairs, measured latency can reasonably predict available bandwidth based on a correlation function derived from observing latency and bandwidth across all node pairs.
منابع مشابه
Autonomic Mechanisms for Service Scaling in Wide-Area Shared Computing Environments
In this paper we present mechanisms and models for building autonomically scalable and resilient services in wide-area shared computing environments. Wide-area shared computing platforms as exemplified by PlanetLab provide a cooperatively shared pool of computing resources. Building services in such environments poses unique challenges due to the fluctuations in available resource capacities an...
متن کاملSurviving Internet Catastrophes
In this paper, we propose a new approach for designing distributed systems to survive Internet catastrophes called informed replication, and demonstrate this approach with the design and evaluation of a cooperative backup system called the Phoenix Recovery Service. Informed replication uses a model of correlated failures to exploit software diversity. The key observation that makes our approach...
متن کاملService Placement in a Shared Wide-Area Platform
Emerging federated computing environments offer attractive platforms to test and deploy global-scale distributed applications. When nodes in these platforms are timeshared among competing applications, available resources vary across nodes and over time. Thus, one open architectural question in such systems is how to map applications to available nodes—that is, how to discover and select resour...
متن کاملCluster Based Bee Algorithm for Virtual Machine Placement in Cloud Data Centre
The utilization of cloud data centres in combination with Virtualization technology has advantages of running more than one virtual machine in a single server. The data centres are a collection of many servers, allocation of VM to Host is known as VM placement. VM placement problem was examined in this paper with focus for maximum utilization of the resources and energy reduction. Switching off...
متن کاملQuiver: Consistent Object Sharing for Edge Services
We present Quiver, a system that coordinates service proxies placed at the “edge” of the Internet to serve distributed clients accessing a service involving mutable objects. Quiver enables these proxies to perform consistent accesses to shared objects, by migrating the objects to proxies performing operations on those objects. These migrations dramatically improve performance when operations in...
متن کامل