“RaaS and Ginseng: The Resource as a Service Cloud”

Fri Feb 12, 2016 3:00 PM

Location: A.V. Williams Building, Room 2120

Assaf Schuster
Professor, Computer Science Department, Technion – Israel Institute of Technology

Cloud providers, such as Amazon EC2, would like to have satisfied clients. Who wouldn’t? However, in order to maximize their marginal profit, they have to fit the clients on as few machines as possible.

One way providers can maximize clients’ satisfaction while making best use of the resources is by renting precious physical memory to those clients who value it the most. But real-world cloud clients are selfish; they will only tell their providers the truth about how much they value memory when it is in the clients’ best interest to do so. Howthen, can cloud providers allocate memory efficiently to those (selfish) clients?

We present Ginseng, the first market-driven cloud system that solves this problem, allocating memory efficiently precisely to those cloud clients who value it the most. Ginseng, built using the KVM hypervisor and libvirt, achieves a 6.2x-15.8x improvement (83 percent-100 percent of the optimum) in aggregate client satisfaction. Time-permitting, we will also describe similar results for the allocation of other resources, currently under submission for publication.

Assaf Schuster is a professor in the Computer Science Department at the Technion – Israel Institute of Technology.

He is an ACM fellow and a world-renowned expert on distributed and scalable data mining, big data technologies analytics and prediction, cybersecurity and system vulnerabilities, privacy preserving, cloud resource management and more.

Schuster has published more than 200 papers in highly selective conferences and journals, some of which have won prestigious awards.

He has consulted leading high tech companies, such as IBM, HP, Microsoft, and Verint. He participated in the bumpy journey of quite a few startups, some of which were successful.

Schuster’s research group is well-known for its contributions to the field of big data and scalable, real-time knowledge discovery in distributed data streams.