The most popular and comprehensive Open Source ECM platform
Cloud Computing: Feds Test the Water of the Cloud
In the face of diminishing budgets, Cloud computing is looking attractive to government agencies. But most agencies are unsure of how to proceed and are looking for guidance. The National Institute of Standards and Technology (NIST) has begun an investigation that seeks to determine how agencies can best take advantage of cloud technology. As part of the investigation, NIST is creating a cloud computing simulation model within their Information Technology Lab (ITL). The code name for the NIST cloud project is Koala.
Dawn Leaf, senior executive of cloud computing at NIST, said that “the objective of the project is to assess and characterize resource allocation algorithms within a public infrastructure-as-a-service cloud model.” Leaf expects to have their first set of results from Koala and be able to make recommendations by early to mid 2011.
NIST’s Project Koala aims to compare resource algorithms for Infrastructure as a Service (IaaS) cloud computing. The goal is to investigate different cloud architectures and to be able to better understand and characterize the complex nature of the cloud. Parallel to the work being done on Koala, government agencies have started to work with major cloud vendors like Google, Microsoft and IBM to help plan the build-out of government cloud infrastructure. The vendors will help answer questions about interoperability, portability and security. Leaf emphasized that while the government will try to move forward in a vendor-neutral fashion, it realized that it needs it work closely with the broad community of vendors to ensure the viability of any cloud plans they develop.
Project Koala will look at a complete range of use cases for deploying the cloud in government agencies. Leaf commented that some use cases for the cloud have clear cut solutions, but many do not. For example, one use case is to move data from an existing government web sites to a public cloud. That kind of data isn’t sensitive so moving it to the cloud is not a reason for concern. On the other hand, government information which is sensitive and needs to be managed in a secure way in a cloud scenario would be best implemented within a private cloud environment. The problem is that most government use cases aren’t always so clear cut; most fall into a gray area in between the two. In those cases, a hybrid cloud model is needed.
The first phase of Koala is to define in vendor-neutral terms exactly what a cloud structure should look like. The model should not prescribe or advocate a particular implementation, but it must be flexible enough to be able to support a solution for a variety of government use cases. Other deliverables of Koala are a roadmap for how to proceed going forward and a gap analysis between existing cloud standards and the roadmap developed by the project.













