Rice professor Eugene Ng leads project to optimize applications
By PATRICK KURP
Suppose you’re the owner of a startup company with a promising future but limited capital, and you need access to more computing services than you can readily afford.
“With cloud computing, you’re buying services from a provider, both hardware and software,” said T.S. Eugene Ng, associate professor of computer science. “You might think of it as a pay-as-you-go computing plan, with the flexibility to buy more or less resources as your needs vary.”
Businesses are turning with increasing frequency to cloud computing as a way to affordably support their operations. Users access cloud-based computer services over a network, typically the Internet.
Ng is leading the “Architecting Performance Sensitive Applications for the Cloud” project with collaborator Sanjay Rao, an associate professor of electrical and computer engineering at Purdue University. The project is supported by the National Science Foundation.
“Our goal is to provide the best solutions for cloud application designers to take cloud application performance and manageability to the next level,” Ng said. “We’re working from the user’s perspective. The cloud is such a rich environment, and there are so many variables, we want to simplify it into a sort of ‘Cloud Computing for Dummies.’ We’re developing new solutions based on performance prediction models, optimization theory and statistical machine learning techniques.”
Cloud computing derives its name from the use of a cloud-shaped symbol for the complicated infrastructure it contains in system diagrams, including databases, Web servers and computation capacity. Cloud computing entrusts remote resources with a user’s data, software and computation, while permitting the user to increase his capabilities without investing in new infrastructure, training new people or licensing new software.
Ng and his partners will also evaluate hybrid cloud computing, the simultaneous use of both public and private computing resources, and the implications of “virtualization technology,” in which the resources of a single physical computer are shared among several virtual computing environments.
“The advantages of cloud computing are cost savings and agility,” Ng said. “An application can potentially scale and move among different cloud environments very quickly. We’re all about efficiency and figuring out the best way to leverage cloud computing.”