Serverless edge computing uses an event-driven model in which Internet-of-Things (IoT) services are run in short-lived, stateless containers only when invoked, leading to significant reduction of resource utilization. However, a cold-start of a container can take up to several seconds which significantly degrades the response time of serverless applications. Container caching can mitigate the cold-start problem at the cost of extra computing resources which violates the spirit of serverless computing. Therefore, we need to balance the cold-start overheads with the extra resource utilization for serverless edge computing. Nevertheless, the diverse ranges of containers lead to different cold-start overheads, resource consumption and invocation frequencies and these characteristics of containers are largely overlooked by existing caching policies. In this paper, we study the request distribution and caching problem for serverless edge computing. We devise an online request distribution algorithm with performance guarantee and present an adaptive caching policy which incorporates container frequency, container size and cold-start time. Via real-system implementation, the superiority of the proposed algorithm is verified by comparing with existing caching policies, including fixed caching and histogram based policies. Our results show that the proposed algorithm reduces both the average response time and cold-start frequency by a factor of 3 compared to current approaches.
Balls-into-bins games for uniform bins are widely used to model randomised load balancing strategies. Recently, balls-into-bins games have been analysed under the assumption that the selection probabilities for bins are not uniformly distributed. These new models are motivated by properties of many peer-to-peer (P2P) networks. In this paper we consider scenarios in which non-uniform selection probabilities help to balance the load among the bins. While previous evaluations try to find strategies for identical bins, we investigate heterogeneous bins where the “capacities” of the bins might differ significantly. We look at the allocation of m balls into n bins of total capacity C where each ball has d random bin choices. For such heterogeneous environments we show that the maximum load remains bounded by lnln(n)/ln(d)+O(1)w.h.p. if the number of balls m equals the total capacity C. Further analytical and simulative results show better bounds and values for the maximum loads in special cases
Although cloud computing has become an important topic over the last couple of years, the development of cloud-specific monitoring systems has been neglected. This is surprising considering their importance for metering services and, thus, being able to charge customers. In this paper we introduce a monitoring architecture that was developed and is currently implemented in the EASI-CLOUDS project. The demands on cloud monitoring systems are manifold. Regular checks of the SLAs and the precise billing of the resource usage, for instance, require the collection and converting of infrastructure readings in short intervals. To ensure the scalability of the whole cloud, the monitoring system must scale well without wasting resources. In our approach, the monitoring data is therefore organized in a distributed and easily scalable tree structure and it is based on the Device Management Specification of the OMA and the DMT Admin Specification of the OSGi. Its core component includes the interface, the root of the tree and extension points for sub trees which are implemented and locally managed by the data suppliers themselves. In spite of the variety and the distribution of the data, their access is generic and location-transparent. Besides simple suppliers of monitoring data, we outline a component that provides the means for storing and preprocessing data. The motivation for this component is that the monitoring system can be adjusted to its subscribers - while it usually is the other way round. In EASI-CLOUDS, the so-called Context Stores aggregate and prepare data for billing and other cloud components.
1 February 2012 in Journal of Parallel and Distributed Computing