Bias on Public Clouds for Big Data
Okay, so admittedly there is a bit of trickery to the headline; the bias we’re talking about forks into two directions. First, the subject of a recent interview is Randy Bias, creator of CloudScale Networks, one of the technical visionaries behind GoGrid and current CEO of CloudScaling.
The second element of “bias” is rooted in a conversation that Randy Bias had with Robert Duffner about the evolution of cloud computing and how public and private cloud models have biased networks of proponents and uses backing them.
While the interview is extensive and covers a range of cloud issues that are particularly relevant to enterprise cloud users, Duffner did ask Randy Bias about what role public clouds might play in high performance computing and what cloud means for former or current grid users.
Bias stated calls cloud computing “high scalability computing” and not “high performance computing” for public clouds, explaining that this means “the non-HPC use cases at the lower end of the grid market already make sense on the public clouds today. If you run the numbers and the cost economics make sense, you should embrace cloud-based grid processing today.”
The CloudScaling CEO noted that “Amazon is building out workload-specific portions of their cloud for high performance computing running on top of cloud. Still, at that very top of the current layer of grid use cases that are HPC, the cost economics for cloud are probably never going to make sense.” To highlight how such a use case might not make sense, Randy Bias points to large institutions like CERN that requires low latency infrastructure for MPI problems.
Duffner’s interview with Bias is hosted at the Windows Azure Team Blog and is part of a larger series of interviews entitled, “Thought Leaders in the Cloud” which features conversations with other notables, including Dell’s Barton George, Oak Leaf Systems owner Roger Jennings, and NASA CTO, Chris Kemp among others.
Full story at Thought Leaders in the Cloud