Log in
As a charitable service-based nonprofit organization (NPO) coordinating individuals, businesses, academia and governments with interests in High Technology, Big Data and Cybersecurity, we bridge the global digital divide by providing supercomputing access, applied research, training, tools and other digital incentives “to empower the underserved and disadvantaged.”

Lunch and Learn - February 2018 - Download Presentation! - - Datacenter Infrastructure for HPC: Site selection & special operational considerations. Also a post mortem on Hurricane Harvey.

08 Mar 2018 11:14 AM | SHPCP (Administrator)

The Society of HPC Professionals

 Lunch & Learn Round Table Discussion Meeting

Datacenter Infrastructure for HPC: Site selection & special operational considerations.
Also a post mortem on Hurricane Harvey.


Datacenter infrastructure continues to exist as a requirement for the bottom of the IT stack in all enterprise applications, if the power is not on the machines will not work. As organizations evolve, HPC specialists continue to think about optimizing the cost of computing, which includes topics such as resource optimization, but also carry weight in conversations around compromises in uptime with mixed reliability infrastructures and build vs. buy.

No matter where an operation is deployed, the site and operational risks must be established to properly deliver your concept of operations. This has implications on where you build, what you build, and when you build. Often times, the hardest part of engineering a datacenter is not making the electrons flow, but matching the infrastructure to the concept of operations of your specific deployment.

This lunch and learn session will begin with an overview of the state of datacenter design as it relates to high performance computing, and mixed reliability architectures. We will then discuss key criteria for building into existing datacenters, along with specifications that are important and often overlooked for spec-built facilities. Hurricane Harvey, and generally Houston, will be analyzed as a case study for datacenter site selection and design. Finally, we will look at Build vs. buy (leasing with colocation) decision making matrices, discussing important pitfalls of both.

Presenter’s Bio:

Chris Hinkle – TRG Datacenters CTO

Chief Technology Officer Chris Hinkle leads datacenter design, operations, and sales engineering for TRG. At prior datacenter companies, Chris was a key contributor in engineering efforts for Fortune 500 and Healthcare clients. He has held prior research oriented positions working on Military/DoD applications at MIT Lincoln Laboratory, and is published in the areas of power systems and sensors. While at MIT, Chris utilized HPC for geospatial analysis based on GIS integrated with LADAR. Chris obtained his BS Electrical Engineering from Virginia Tech, Magna Cum Laude, and his MBA from Emory University.