Abstract

Infrastructure-as-a-service (IaaS) Clouds concurrently accommodate diverse sets of user requests, requiring an efficient strategy for storing and retrieving virtual machine images (VMIs) at a large scale. The VMI storage management require dealing with multiple VMIs, typically in the magnitude of gigabytes, which entails VMI sprawl issues hindering the elastic resource management and provisioning. Nevertheless, existing techniques to facilitate VMI management overlook VMI semantics (i.e at the level of base image and software packages) with either restricted possibility to identify and extract reusable functionalities or with higher VMI publish and retrieval overheads. In this paper, we design, implement and evaluate Expelliarmus, a novel VMI management system that helps to minimize storage, publish and retrieval overheads. To achieve this goal, Expelliarmus incorporates three complementary features. First, it makes use of VMIs modelled as semantic graphs to expedite the similarity computation between multiple VMIs. Second, Expelliarmus provides a semantic aware VMI decomposition and base image selection to extract and store non-redundant base image and software packages. Third, Expelliarmus can also assemble VMIs based on the required software packages upon user request. We evaluate Expelliarmus through a representative set of synthetic Cloud VMIs on the real test-bed. Experimental results show that our semantic-centric approach is able to optimize repository size by 2.2-16 times compared to state-of-the-art systems (e.g. IBM's Mirage and Hemera) with significant VMI publish and retrieval performance improvement.

Highlights

  • The evolving Cloud architecture [4], [7], [17], [25], [28] requires efficient and scalable on-demand provisioning and management of computing services over a federated and heterogeneous infrastructures

  • We define the semantic graph of a virtual machine images (VMIs) I as a directed cyclic graph GI = (VI, EI ), where VI = base image subgraph GI (BI) ∪ package subgraph GI (P S) ∪ DS represents the set of vertices including the base image, primary and dependency packages, and EI ⊆ VI × VI is the set of edges, where a direct edge e = (v, v′) ∈ EI denotes a dependency of the base image, primary package, or dependency package v on v′

  • In the lack of any public VMI management benchmark, we evaluate our approach using a synthetic VMI set based on the Ubuntu Linux distribution with software packages recognized by package management tools

Read more

Summary

INTRODUCTION

The evolving Cloud architecture [4], [7], [17], [25], [28] requires efficient and scalable on-demand provisioning and management of computing services over a federated and heterogeneous infrastructures. To solve VMI management challenges such as sprawl, prior research in this domain primarily focused on leveraging VMI deduplication [14], [16], [18] and caching [11], [19], [22], [29] by identifying similar byte segments [10], [12], [20] Such techniques optimize the VMI storage and reduce redundant content by up to 80%, but limit the benefits of the virtualization technology, such as exploiting stronger isolation between software packages, and keeping a provenance record of changes and reusable functionality in the VMI at a semantic level [23].

RELATED WORK
VMI SEMANTIC MODEL
VMI semantic graph
VMI attributes
Semantic compatibility
VMI master graph
Architecture overview
VMI semantic analyzer
VMI decomposer
VMI assembler
IMPLEMENTATION
Experimental setup
VMI repository optimization
VMI publishing and retrieval
CONCLUSION
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call