NA3.3.2 - EMI Cloud Article Website October 2011

Version 1

Grid computing gained a prominent position through the concept of decentralized resource sharing in VOs solving complex scientific use cases executed by geographically dispersed user communities. One of the challenges raised by e-Science applications is to have urgent resource access and expansion, lower resource downtimes, more administrative control, and less maintenance costs. The emerging virtualization techniques leading to cloud computing technologies aim to deliver solutions to these challenges. The benefits of on-demand provisioning, offering cost of almost zero maintenance, allowing more control and flexibility to end-users.

Cloud computing is considered to be a new paradigm also affecting the European Distributed Computing Infrastructure (DCI) landscape by delivering end users with an access to the compute, software, storage, and data services without direct knowledge of their location and configuration. As a consequence, also EMI needs to investigate in potential technical innovations that can be derived from this paradigm while at the same continue to deliver mature production products to DCIs.

The US National Institute of Standards and Technology (NIST) activities define cloud computing as follows: Cloud computing is a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction.

From a more specific implementation perspective and theory of distributed systems, cloud computing is not an innovation, in fact, it has a basis on the effective use of existing Internet protocols, virtualization, SOA, and utility computing models. Hence, the technology innovation can be considered to be complementary to the EMI products and its technology solutions. More recently, EMI explores various cloud options in order to understand the effective use of underlying resources using the cloud paradigm with EMI products to enable a secure, on-demand, and cost effective infrastructure with compute, data, and storage resources.

Members of the EMI project investigate the use of the cloud paradigm in two major ways. First, they analyse what changes are needed in order to run EMI products within virtual appliances running in virtualized environments. In general, EMI products work in such environments, while several new challenges are dynamic configuration of such products, including the challenge of setting up an EMI ecosystem across different EMI virtual appliance instances (i.e. communication aspects). Second, EMI members analyse which EMI products might be suitable for their re-use on the virtual machine management layer in order to be re-used from cloud technology providers.

Finally, EMI needs to state that the focus of the project remains on delivering mature production products in order to enable scientific applications with todays DCIs. Although cloud seems to be a very promising new paradigm, EMI is not a dedicated cloud project and as such is only exploring and investigating potential solutions while cloud products need to be provided by other collaborative projects. Nevertheless, some of the EMI products proven their suitability in distributed systems over one decade and might become a technology of choice for cloud technology provider.
Edit | Attach | Watch | Print version | History: r1 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r1 - 2011-10-07 - MorrisRiedelExCern
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    EMI All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2021 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback