As a vendor of management tools for EDA farms, we provide a suite of products to maximize the utilization of your computing resources. Such resources include the compute farms owned by our customers, which are always finite. Our products keep all users as productive as possible within the infrastructure the customers already have. By far the most expensive resource is the time based software licenses (TBL). We have tool to detect bottlenecks in the workflow caused by resource saturation and we can recommend solutions to eliminate them, typically involving the provisioning of additional hardware or software, which cannot generally be done in real time.
Enter the "Cloud". According to a widely accepted technical report from University of California, Berkeley, Above the Clouds: A Berkeley View of Cloud Computing the cloud is a business model that offers:
- The illusion of infinite computing resources available on demand, thereby eliminating the need for Cloud Computing users to plan far ahead for provisioning.
- The elimination of an up-front commitment by Cloud users, thereby allowing companies to start small and increase hardware resources only when there is an increase in their needs.
- The ability to pay for use of computing resources on a short-term basis as needed (e.g., processors by the hour and storage by the day) and release them as needed, thereby rewarding conservation by letting machines and storage go when they are no longer useful.
The litmus test for an EDA farm is to have enough resources for steady state demand plus the ability to accommodate peaks without a major disruption in the low priority projects. Our WorkloadAnalyzer can detect bottlenecks and suggests precisely what to do, like acquiring more licenses or more servers, specifying the type of servers. Clouds however are supposed to dynamically supply resources when needed. Considering that in the cloud everything has a price per use, the scheduler must consider the cost of the resources rather than just their utilization. A cloud in EDA must schedule for minimum cost, not just maximum utilization.
This probably could be done, but how do we create "the illusion of infinite licenses"? The EDA ISVs control this pricing model based on TBL. If the license manager can deliver pay-per-use licenses via subscriptions, then a production EDA - private, public, perhaps hybrid - cloud will be possible.
Olivier Coudert's Blog reports about some pioneering EDA clouds already in operation.
The big EDA companies have been slowly espousing the idea of making their tools available in the cloud. Cadence set up a (private) cloud offering a while back, but it has never been successful. Synopsys announced in March that it would provide a cloud computing solution for VCS simulation with AWS (Amazon Web Services)...
...startups have been testing the water. Xuropa hosts demos and a CRM platform in the cloud, and counts Cadence and Synopsys as customers. Plunify proposes FPGA synthesis with multiple runs in the cloud. Nimbic (formerly known as Physware) uses the cloud to develop and deploy its tool. Same for Tabula Stylus software cloud platform. Xuropa, Plunify, and Nimbic are now making claims about moving other EDA applications in the cloud
We are watching closely. We have been managing EDA farms for sixteen years and we know any cloud solution faces a tremendous challenge of satisfying simultaneously the needs fo the EDA users (elasticity in particular) an the demands of the ISVs. In particular, we are waiting to see which pay-per-use pricing model will be most adopted.
