editor's blog
Subscribe Now

An Almost-Cloudy San Diego Day

Not long ago we looked at how EDA is shaping up in the cloud, including work that Synopsys has been doing to make VCS available for bursty relief usage. I was fortunate enough to attend a demo session to show how what has heretofore been an interesting theoretical discussion could be made concrete.

Synopsys spent a lot of effort on cloud computing at DAC this year, including a cloud partners booth. Various names, both obvious and some not so, were in the booth: Amazon, NetAp, Cisco, CloudPassage, Univa, Platform Computing, Xuropa, and EVE. Most of these guys provide a variety of services to bolster the Infrastructure as a service (IaaS), Platform as a service (PaaS), or Software as a service (SaaS) layers of any EDA offering. They appear to have very specific EDA (not just Synopsys) messaging.

In the demo session itself, we saw a bit more of Synopsys’s VCS specifics. As mentioned before, their setup involves a master node and multiple work nodes. Each cloud-computing instance (CCI) node consists of an 8-core Nehalem machine with 23 GB of memory for the 8 cores, along with a single VCS license. You can then requisition clusters consisting of multiples of 8 nodes, sizing up to hundreds of nodes (although they want advance notice on really large requests for now so they can set that up with Amazon; they don’t currently have them lying around because that costs money and there’s not that much demand yet).

They showed the scripts used to get things set up. The bring-up process lasts about a half an hour (they didn’t try to run that live), which might sound like a long time until you realize that, in that time, you’ve gone from nothing to, potentially, a multi-hundred-server compute farm.

I’d like to report on how they then went into the cloud and ran an example OpenSPARC simulation. That was the plan. But the unthinkable happened. And I totally felt for the guy running the demo. I mean, it’s the nightmare scenario for any of you (and me) who have done demos: you go to where the files are all ready to be uploaded and run… and… they’re gone. Completely gone. Like, the folder isn’t even there.

Turns out that a hard drive died on the Synopsys campus. The drive housing the project they were going to demonstrate. File it under “W” for “WT…” well, you know how that one ends. Yeah, crazy. So there wasn’t time to reconstruct it on another drive and start again. So we’ll have to report later when we see things actually happening in the cloud.

Leave a Reply

featured blogs
Nov 30, 2023
No one wants to waste unnecessary time in the model creation phase when using a modeling software. Rather than expect users to spend time trawling for published data and tediously model equipment items one by one from scratch, modeling software tends to include pre-configured...
Nov 27, 2023
See how we're harnessing generative AI throughout our suite of EDA tools with Synopsys.AI Copilot, the world's first GenAI capability for chip design.The post Meet Synopsys.ai Copilot, Industry's First GenAI Capability for Chip Design appeared first on Chip Design....
Nov 6, 2023
Suffice it to say that everyone and everything in these images was shot in-camera underwater, and that the results truly are haunting....

featured video

Dramatically Improve PPA and Productivity with Generative AI

Sponsored by Cadence Design Systems

Discover how you can quickly optimize flows for many blocks concurrently and use that knowledge for your next design. The Cadence Cerebrus Intelligent Chip Explorer is a revolutionary, AI-driven, automated approach to chip design flow optimization. Block engineers specify the design goals, and generative AI features within Cadence Cerebrus Explorer will intelligently optimize the design to meet the power, performance, and area (PPA) goals in a completely automated way.

Click here for more information

featured paper

Power and Performance Analysis of FIR Filters and FFTs on Intel Agilex® 7 FPGAs

Sponsored by Intel

Learn about the Future of Intel Programmable Solutions Group at intel.com/leap. The power and performance efficiency of digital signal processing (DSP) workloads play a significant role in the evolution of modern-day technology. Compare benchmarks of finite impulse response (FIR) filters and fast Fourier transform (FFT) designs on Intel Agilex® 7 FPGAs to publicly available results from AMD’s Versal* FPGAs and artificial intelligence engines.

Read more

featured chalk talk

Introduction to Bare Metal AVR Programming
Sponsored by Mouser Electronics and Microchip
Bare metal AVR programming is a great way to write code that is compact, efficient, and easy to maintain. In this episode of Chalk Talk, Ross Satchell from Microchip and I dig into the details of bare metal AVR programming. They take a closer look at the steps involved in this kind of programming, how bare metal compares with other embedded programming options and how you can get started using bare metal AVR programming in your next design.
Jan 25, 2023
37,022 views