The Amp Hour Electronics Podcast

Chris Gammell and Dave Jones' voices span the chasm of thousands of miles each and every week to speak to each other and industry experts about where the field of electronics is moving. Whether it be a late breaking story about a large semiconductor manufacturer, a new piece of must-have test equipment or just talking through recent issues with their circuit designs, Chris and Dave try to make electronics more accessible for the listeners. Most importantly, they try and make the field of electronics more fun. Guests range from advanced hobbyists working on exciting new projects up through C-level executives at a variety of relevant and innovative companies. Tune in to learn more about electronics and then join the conversation! Visit The Amp Hour website for our back catalog of 150+ episodes.

https://theamphour.com

subscribe
share






#357 – An Interview with Rick Altherr


Welcome, Rick Altherr (@kc8apf)

  • Rick works on the platform team at Google, designing firmware and software for the servers that go into hyperscale datacenters. He does not discuss specific Google stuff.
  • Chris was surprised when Rick walked into a meetup carrying the Zaius server under his arm.
  • The standard 1U rack size is 19″. This is the first of many standardizations that have happened over the years.
  • You can purchase a slot, a rack, an entire cage or a building. Each comes with an increasing amount of logistical questions and potential for customization.

  • Engineers are following “Scale out” architecture these days.
  • Kubernetes
  • Gluster
  • Reed-solomon codes
  • MTBF of hardware

    Your [server hardware] product is not truly shipped until it has caught fire at a datacenter

  • These servers are part of the Open Compute project, which includes many large companies in the tech space.
  • Two of the popular designs are the Leopard and the Zaius
  • The Zaius has 2 Power9 processors, 48V rails, point of load converters that take the voltage down to 0.8V, and slots for 32 sticks of RAM. You can view all the design files on GitHub.
  • PUE – power usage effectiveness
  • Hot isle/cold isle
  • Lights out management type system
    • Open Stack
    • Kubernetes
    • MaaS
  • OpenBMC – Chassis manager
  • NCSI – connect the ethernet
  • IMPI – controlled by DMTF
  • Project olympus motherboard microsoft
  • You can just…BUY…these servers from the Open Compute project.
  • Go on a 360 degree tour of the datacenters at Google
  • Read more about Google’s datacenters here.
  • Decisions are made on metrics like “People per megawatt” (how much management is required) and “Build vs buy” (is there a solution already in the marketplace).
  • Finding a job: search for terms like “Datacenter”, “Platform”, or ping Rick via DM for mentoring and/or practice interviewing
  • Rick gave a great talk at HDDG about monitoring engines called “Knock knock: Add more fuel“
  • You can find Rick on Twitter or GitHub

Occasional barks during the show provided by Rick’s new puppy!

I couldn't dream up these poses pic.twitter.com/FP0brIFRYK

— Rick Altherr (@kc8apf) August 27, 2017


fyyd: Podcast Search Engine
share








 August 29, 2017  1h38m