What Good Did It Do? Quite a Lot, and Quite the Story

A typical show opening, but we’re missing Henry – he was away packing for his big move from his long-time home base in Minnesota to his new bunker in Las Cruces. He’s packing up all of his memorabilia, like his first punch card sets, his core memory collection and his home terminal, lovingly wrapping them up for the long trip south.

News Alert

NEWS ALERT:  we interrupt this show blog for a special announcement. We’re having a contest! This is the digital version of the “18th caller” contest. The 18th emailer to RFHPC, starting now, will receive what Dan might describe as a fantastic prize from us at Radio Free HPC. Send in your email entries now, and listen to the show for details. This might take a couple of weeks, it might take a couple of years, but we’re nothing if not good with small numbers and large units. And that 18th lucky emailer deserves a prize.

A Cool New Book Project

We have a special guest today, David Barkai, a 50-year veteran of HPC. David has worked in a wide variety of positions at NASA, Intel, Cray, SGI and others. His project now is writing a book to chronicle the last 50 years in HPC told from the perspective of those who were there.

The main emphasis in the book is examining the good that HPC has done in the world, which is quite the story. He’s looking at the applications that have changed the world, from weather forecasting to safer and quieter cars and so on, and the system architectures that have made them possible in a decade-by-decade tour of HPC development.

  • 1970s:  Vector Processors
  • 1980s:  Multiprocessors
  • 1990s:  Massive Parallelism
  • 2000s:  Clusters and Accelerators
  • 2010s and beyond:  HPC and AI/Cloud

As David says in our interview, the top HPC systems have advanced 10-15 times faster than Moore’s Law, which is astounding. In the book he goes into detail about how the industry drove performance at such a dizzying pace. He’s still writing away and is interested in hearing about your HPC journey to help fill out the book. You can reach him here.

Reasons Why No One Should Ever Be Online. Ever.

Henry is practicing what he preaches and is not online right now as he waits to the internet company to get him connected again.

Catch of the Week:

JessiGoogle’s head of Quantum computing hardware resigns. John Martinis resigned after being assigned to an advisory role in the company.

Shahin:  Discusses a paper on Coarse Grain Reconfigurable Architecture (CGRA) as a way of having performance and programmability closer to the metal. Its survey of what’s been going on in this area with FPGAs is great and may also point to what we can expect in future supercomputers.

Dan:  Subs for Henry with a horrible impression for a first catch. But his real catch is a plug for the Radio Free HPC Studio Products production of “Charles Babbage, his life and times.”  It’s an impressive production with chills, thrills and plenty of action. Don’t put it off, listen now. All those reviews are great for a reason!

Join us!

* Download the MP3 
* Sign up for the insideHPC Newsletter
* Follow us on Twitter
Subscribe on Spotify 
Subscribe on Google Play 
Subscribe on iTunes 
RSS Feed
* eMail us

Leave a Reply

Your email address will not be published. Required fields are marked *