Maximize Bandwidth in your Massively Parallel AI SoCs?

Artificial Intelligence is one of the most talked about topics on the conference circuit this year and I don’t expect that to change anytime soon. AI is also one of the trending topics on SemiWiki with organic search bringing us a wealth of new viewers. You may also have noticed that AI is a hot topic for webinars like the one I am writing about now.

We have been working with NetSpeed for 3 years now and have published blogs covering a wide range of topics. You can see their landing page here. NetSpeed has done some of the best and most widely viewed webinars that we have been involved with and I expect this one will be the same.

How do you maximize bandwidth in your massively parallel AI SoCs

Tue, Jul 24, 2018 8:30 AM – 9:00 AM PDT

When designing a SoC for AI applications, you are faced with a system using 1000’s of cores in a massively parallel architecture. Performance, bandwidth and quality of service (QoS) are critical requirements and the challenges of meeting them are very different for these SoCs used for AI. This webinar, in 30-minutes brings out the challenges and the solutions that has empowered multiple leaders in the AI space.

John Bainbridge, Principle Application Architect, NetSpeed Systems, will be presenting. Before joining NetSpeed John worked for Qualcomm on the SnapDragon chips so John knows SoCs, absolutely.

I have an advanced copy of the slides and they are definitely worth a look. Here is a quick outline:

Breaking down the AI workflow:

• How it happens
• What matters
• Critical use cases
• SoC Data FlowArchitectural Challenges:
• Large number of cores
• Extremely high bandwidth
• Peer-peer traffic and multicast
• Sophisticated QoSBottom line: Traditional approaches are inadequate for AI SoCs.

  John then goes into the NetSpeed approach, technology, and QoS support. This is why I like webinars, you get to hear it from and interact with the experts. Not as good as live but definitely the next best thing. Register even if you can make the live event so you automatically get a link to the replay. I hope to see you there!

  About NetSpeed
  NetSpeed Systems provides scalable, coherent on-chip network IPs to SoC designers for a wide range of markets from mobile to high-performance computing and networking. NetSpeed’s on-chip network platform delivers significant time-to-market advantages through a system-level approach, a high level of user-driven automation and state-of-the-art algorithms. NetSpeed Systems was founded in 2011 and is led by seasoned executives from the semiconductor and networking industries. The company is funded by top-tier investors from Silicon Valley. It is based in San Jose, California and has additional research and development facilities in Asia. For more information, visit www.netspeedsystems.com.