Published: 2013 Juni
Buchtitel: FORECAST 2013
Verlag: Open Datacenter Alliance
Enterprise IT operators derive three main benefits from the adoption of cloud technology: 1. Resource “arbitrage” 2. Relatively small units for resource allocation 3. Metered resources In the financial world arbitrage involves the simultaneous negotiation of the same instrument in different markets with the goal of profiting from price differences. In a cloud environment an IT operator has a choice of running workloads in different geographic locations or from services delivered by a variety of providers. In organizations using physical allocation, a new server represents a three - year commitment or longer, and therefore budgeting and procurement are commensurately heavy processes. Securing these services through an internal or external cloud provider is potentially orders of magnitude faster. Cloud resources are delivered through a service model where the service’s performance and cost are specified contractually through a service level agreement. This environment encourages a data driven approach to resource management, allowing for much agile IT processes than those seen under more traditional budgeting models. Most cloud based applications today run in virtualized environments. A necessary condition for realizing the benefits is a capability for moving workloads across machines. These machines can be part of a private cloud or in a public cloud, and with relocating workloads there is a need to move virtual machines within and across different clouds. This capability is described in an Open Data Center Alliance (ODCA) usage model document.
A team led by T-Systems Telekom Innovation Laboratories, the FZI research team from the University of Karlsruhe and supported by Intel Corporation carried out a proof of concept (PoC) project to implement the usages described in the document, described in this report.
In context of the ODCA VM Interoperability Usage Model mentioned, an initial selection
was made of commonly found products in the corporate cloud environment, and a set of
appropriate experiments defined to test
them against the Usage Model.
shown in this document demonstrate how closely these potential systems aligned to the
Usage Model at the time of the PoC tests towards demonstrat
ing Cloud technology
Later iterations of testing may expand the scope of products as far
for a PoC test, and as software versions evolve.
Therefore the results shown represent a
baseline for cloud expectations and current cloud technology state-of-the-art, which should significantly improve for every subsequent test which the reader
may wish to initiate.
Download: Media:VM Interop PoC White Paper.pdf