What I learned from years in the cloud

What I learned from years in the cloud

As part of the HPC User Forum, Jason Stowe discussed some of key benefits that we have found after years of making customers successful doing their HPC and large computation in the cloud. In this presentation, he breaks it down to a number of key learnings. And, for the time pressed, he touches on the summary early in the presentation. You can see a video of the talk here  or watch below.  ...
Great Financial Services cloud talk at HPC User Forum

Great Financial Services cloud talk at HPC User Forum

Jeffrey Smart from AIG did a great presentation at the HPC User Forum in Tucson, AZ April 11-13. He described his experience in moving some of the time critical AIG Risk Analysis workload from finite internal resources to the full flexibility of the cloud. His talk includes discussing how they were able to expand their workflow and offering to the business and some comments about the challenges in going cloud. You can see the video...
Jason Stowe, Cycle Computing CEO, presented at the 2016 Stanford HPC Conference

Jason Stowe, Cycle Computing CEO, presented at the 2016 Stanford HPC Conference

On February 24th, Jason helped to open the 2016 HPC Advisory Council event at Stanford University. Jason’s talked about how nine years of Cloud HPC experiences has taught us a number of lessons about things like security, scale, workload planning, cost models, multi-users, multi-cloud, and the difference between running one job versus multiple workflows in the cloud. InsideHPC captured the talk on video so that you can see it in its entirety or just watch it below  ...
HPC on Cloud: Results from the Record-breaking 156,000 Core MegaRun, and Insights to On-demand Clusters for Manufacturing Production Workloads

HPC on Cloud: Results from the Record-breaking 156,000 Core MegaRun, and Insights to On-demand Clusters for Manufacturing Production Workloads

In case you missed this great session at the AWS re:Invent 2014 conference, Cycle CEO Jason Stowe hosted HGST (a Western Digital company), and University of Southern California (USC) to talk about HPC on the Cloud. See below for full session video, and abstract description. BDT311 – HPC on AWS: Results from the Record-breaking 156,000 Core MegaRun, and Insights to On-demand Clusters for Manufacturing Production Workloads Not only did the 156,000+ core run (nicknamed the MegaRun) on Amazon EC2 break industry records for size, scale, and power, but it also delivered real-world results. University of Southern California (USC) ran the high-performance computing job in the Cloud to evaluate over 220,000 compounds and build a better organic solar cell. In this session, USC’s Patrick Saris provides an update on the six promising compounds he found that he is now synthesizing in laboratories for his clean energy project. Saris discusses the implementation of and lessons learned in running a cluster in eight AWS regions worldwide, with highlights from Cycle Computing’s project Jupiter, a low-overhead cloud scheduler and workload manager. This session also looks at how the MegaRun was financially achievable using the Amazon EC2 Spot Instance market, including an in-depth discussion on leveraging Amazon EC2 Spot Instances to reduce costs and maximize value, while maintaining the needed flexibility, and agility that AWS is known for. After a year of production workloads on AWS, HGST (a Western Digital company), has zeroed in on understanding how to create on-demand clusters to maximize value on AWS. HGST’s David Hinz outlines the company’s successes in addressing the company’s changes in operations, culture, and behavior to this new vision of on-demand clusters. In addition, the session will provide insights...
Video: Jason Stowe Outlines the New Enterprise Workload of Cloud HPC in  AWS re:Invent 2014 Interview

Video: Jason Stowe Outlines the New Enterprise Workload of Cloud HPC in AWS re:Invent 2014 Interview

Recorded Live from the re:Invent 2014 (read our re:Invent blog) show floor in Las Vegas, Cycle Computing CEO Jason Stowe is interviewed by The Cube. In this in-depth conversation, Stowe covers everything from Cycle’s latest Enterprise-speed Cloud cluster it enabled for HGST (a Western Digital company) in November 2014 (read about the record-setting Enterprise Cloud HPC run here), to the company’s record-setting 156,000 core run conducted for researchers at University of Southern California (USC). But all of this isn’t done to set records and build bigger Cloud clusters, Stowe outlines – it’s all in the spirit of better arming researchers & engineers with the powerful tools they need to help them do their jobs better. “The idea that you can just borrow 10-20,000 cores and give [those cores]  back when you’re done is just crazy. This all leads to enabling researchers & engineers to ask the right question – at a scale [beyond] their own internal clusters, to allow them to get better answers, faster.” – Jason Stowe, Cycle Computing CEO Click below to watch and listen to the full...