Report

Teaser, summary, work performed and final results

Periodic Reporting for period 2 - ENTICE (dEcentralized repositories for traNsparent and efficienT vIrtual maChine opErations)

Teaser

What is the problem/issue being addressed?Virtualization is a key technology in Cloud computing that allows users to run multiple virtual machines (VM) with their own application environment on top of physical hardware. Essentially VMs are virtual slices of large, physical...

Summary

What is the problem/issue being addressed?
Virtualization is a key technology in Cloud computing that allows users to run multiple virtual machines (VM) with their own application environment on top of physical hardware. Essentially VMs are virtual slices of large, physical servers. Virtualization enables scaling up and down of applications by elastic on-demand provisioning of VMs in response to their variable load to achieve increased utilisation efficiency at a lower operational cost, while guaranteeing the desired level of Quality of Service (QoS, such as response time) to the end-users. Typically, VMs are created using provider-specific templates (so-called VM images) that are stored in proprietary repositories, leading to provider lock-in and hampering portability or simultaneous usage of multiple federated Clouds.

In this context, optimisation at the level of the VM images is needed both by the applications and by the underlying Cloud providers for improved resource usage, operational costs, elasticity, storage use, and other desired QoS- related features. We identify in this project five critical barriers that prevent many users from industry, business and academia to effectively use Cloud resources and virtualized environments for their computing and data processing needs: (i) manual, error-prone and time consuming VM image creation, (ii) monolithic VM images with large deployment and migration overheads, (iii) proprietary unoptimised VM repositories, (iv) inelastic resource provisioning, and (v) lack of information to support effective VM image optimisation.


Why is it important for society?
In European society, there are a large variety of industrial applications and users that can seriously benefit from the ENTICE environment, such as: Cloud providers, application developers and most importantly Cloud users.
Societal benefits of ENTICE will be the streamlined, automated and intuitive deployment process with simple, optimised, and predictable performance that helps to save time and effort to get customers cloud
ready. Hence, the ENTICE environment will provide substantial advantages to many potential costumers in the SaaS domain such as lower capital expenditures, no need to manage upgrades and patches, and enterprise scalability.

With the help of the lightweight ENTICE distributed VM repository, SaaS and IaaS-based applications and infrastructures will gain increased availability and elasticity. Thus, ENTICE and its technology brings unique Cloud advancements which go far beyond any other existing service or environment available in Europe and word-wide. At least one use case (DEIMOS EOD) will be deployed across the world with the help of ENTICE. For this to happen, ENTICE will be integrated with other non-European IaaS providers like FutureGrid and Amazon EC2.

ENTICE will ultimately be capable of receiving unmodified and functionally complete VM images from users, and transparently tailor and optimise them for specific Cloud infrastructures with respect to their size, configuration, and geographical distribution, such that they are loaded, delivered, and executed faster and with improved QoS compared to their current performances.

Work performed

By leveraging the scalability properties of the distributed Cloud storage infrastructures and exploring sophisticated methods for efficient operational management of federated VMI repositories, we have achieved the following goals:
• Reducing VMI creation time for both expert and unexperienced users,
• Automated creation of optimized VMIs without redundant software packages,
• VMI decomposition in smaller reusable parts,
• Advanced middleware for optimized VMI management,
• Generic multi-objective optimization framework for VMI management,
• Multi-criteria based initial VMI upload and redistribution,
• Multi-objective fragment discovery,
• Novel storage scalability model,
• Online VMI composition utilising of VM templates,
• Interoperability support for various types of VMI formats,
• Advance knowledge management through the use of semantics and a knowledge base,
• Pareto-based Service Level Agreement specification and management.

As a whole, the ENTICE environment can be used to setup and maintain sophisticated system for management of multiple geographically distributed VMI repositories, therefore enabling streamlined adaptation of the next generation Cloud technologies. The impact of the developed technologies has been evaluated with the help of a user interest group acting as reviewers.

The main results of ENTICE include an optimised image repository, an automated technique to identify image fragments across distributed repositories, a multi objective heuristic algorithm for optimised virtual macihne image distribution, a tool for synthesis and optimisation of applications, all targeting service providers. Moreover, we have developed a software functionality design tool targeting application developers and an Image portal to be used by cloud providers.

Final results

The ENTICE project has progressed beyond the state-of-the art and provided the following innovation:

1. Multi-objective Middleware for Distributed VM image repositories in Federated Cloud Environment, which has been designed to provide an easy to use interface capable of receiving unmodified and functionally complete VM images from its users, and transparently distribute them to a specific Cloud infrastructure in a federation with respect to their size, configuration, and geographical distribution, such that they are loaded and delivered faster and with improved reduced financial cost compared to existing solutions.

2. Knowledge Base for Automated Decision Making in Virtual Machine Images Distribution, Analysis and Synthesis, which encompasses a knowledge base built on top of Open Source technologies, such as Jena Fuseki, and reasoners, such as Pellet, and standards, such as OWL/RDF. The ENTICE Knowledge Base can be used along with the ENTICE environment or as a standalone service to assist software engineers to automate decision making in complex situations, for example where there is the need for multi-criteria optimisation.

3. Middleware for Virtual Machine Image Synthesis and Size Optimization, which encompasses a middleware for automated synthesis of highly-optimised VM images for efficient deployment and execution of user applications and/or services. The middleware provides two approaches: (i) start with an already existing VM image and eliminate parts not necessary for the user indicated functionality, and (ii) synthesise VM images from scratch by optimising recipe descriptions. The proposed middleware focuses on services that are widely hosted in virtualised environments, but delivered as a monolithic block of multitude of sometimes vaguely related functionalities.

Whereas ENTICE may not have an immediate impact on socially important factors, ENTICE is part of a technology that aims to pave the way for the development and exploitation of Cloud computing benefiting science, economy and society. ENTICE holds significant potential to contribute to the delivery of public value by increasing operational efficiency and responding faster to constituent needs also for applications and technology that provide and build solutions for a better world.

Website & more info

More info: http://www.entice-project.eu/.