Here's a summary of a brief email interview I conducted with Abhishek Chandra, an assistant professor at the University of Minnesota, Twin Cities, and a program committee member/speaker for the HotCloud workshop taking place in San Diego next week (yesterday I took a spin through some of the interesting papers being presented there):
What's your sense of how hot a topic cloud computing is across the academic and corporate research landscape today?
I don't really have quantitative data, but there have been several recent workshops, as well as sessions and publications in several systems conferences such as OSDI, USENIX ATC, HPDC, etc. It's a fairly hot topic right now - to a large extent it was driven by the industry but has become very popular in academia as well.
What do you think cloud computing's biggest strengths and shortcomings are today?
The biggest strength is the outsourcing and pay-as-you-go model that relieves most customers from infrastructure deployment and management costs. The biggest shortcoming is its use for data-intensive applications/services, both from efficiency as well as privacy/security perspective.
I attended the Interop conference in Vegas a few weeks back and they had a special Cloud event running concurrently. I sat in on one session that featured a couple of early cloud adopters and one of the guys said something along the lines of "Even I'm sick of the term cloud computing." Is cloud getting overhyped?
That's true to some extent. Many of the ideas being applied in clouds have already been there - virtualization, hosting platforms, Grid computing, etc. Clouds are largely an integration of some of these ideas, however, this integration has raised some interesting new challenges, and the business model is also something that is driving the interest.
I took a look at your papers being presented at HotCloud and summed up the Nebulas one in a previous post, but what about the one on Virtual Putty ?
This paper focuses on virtualized cloud platforms, where applications are encapsulated with virtual machines hosted on physical servers. We argue that cloud providers can significantly lower operational costs, and improve hosted application performance, by accounting for affinities and conflicts between co-placed virtual machines. We show how these affinities can be inferred using location-independent VM characterizations called virtual footprints, and then show how migration/placement policies based on these affinities can be used to reshape the physical footprint of a VM-its physical resource consumption-to achieve higher VM consolidation and application performance in a cloud environment. We also identify three general principles for minimizing a virtual machine's physical footprint, and discuss challenges in applying these principles in practice.
The Leader in Network Knowledge
Post new comment