The QTS Experience

Somewhere along the way “diesel” turned into a dirty word. Jim McDonald’s spent 30 years on the generators that keep 911, hospitals, and your data alive — and he’s a little tired of playing the villain. We get into what’s real, what’s myth, and why he still thinks everybody wins.

Chapters:

00:00:00 Welcome, and the “luminary chair”

00:01:24 From the University of Delaware to founding MIRATECH (1992)

00:04:16 The first data center (2006) and the “internet hotel”

00:06:52 What a backup generator does, and why a data center needs it

00:10:16 The “nines” of reliability and the 7-second diesel start

00:13:27 What “five nines” actually means in downtime

00:16:42 How often the generators really run (mostly testing)

00:19:44 Diesel’s “Mufasa moment”: how a word got loaded

00:24:00 “Dirty diesel” — where the myth comes from

00:25:07 The ratepayer pledge and the peak-power problem

00:27:06 Why diesel: 48 hours of fuel on site

00:30:15 Behind-the-meter and microgrids, explained

00:36:14 202(c) orders, Ashburn, and grid-vs-on-site emissions

00:48:14 PJM, the load-duration curve, and demand response

01:02:10 EPA Tier 2 to Tier 4 and a “new spirit of cooperation”

01:14:57 Energy storage: heating bricks, ice, and synthetic DNA

01:30:05 Small modular reactors and the nuclear question

01:33:24 Three daughters and “nuclear is super safe”

01:40:44 What to watch in the next 18–24 months

01:43:13 The takeaway: open minds, and everybody wins

References and Links:

Jim McDonald (guest), MIRATECH — LinkedIn: linkedin.com/in/jim-mcdonald-80229223

MIRATECH (guest’s company): miratechcorp.com

David McCall (host): linkedin.com/in/davidnmccall

QTS: q.com

Ratepayer Protection Pledge (White House, March 2026): whitehouse.gov — Ratepayer Protection Pledge

PJM Interconnection (regional grid operator referenced re: load-duration curve / demand response): pjm.com

Book referenced — “The Mysterious Case of Rudolf Diesel” by Douglas Brunt (2023).

Plant Vogtle (Georgia Power / Southern Company) — the Georgia nuclear plant Jim worked near during construction.

Michael Shellenberger nuclear talk (~10–15 years ago, “several million views”)

Thank you for visiting our channel, be sure to Like, Comment, and Subscribe!

For more information, go to https://www.QTSDataCenters.com

Learn More about QTS Data Centers on the QTS YouTube Channel: https://www.youtube.com/users/qtsdatacenters

Think you have a great topic for a podcast or would like to schedule an interview with Dave? Send in your requests to podcast@q.com

We’re all chasing the same thing: safe, affordable, local power. Rahul Kar thinks we keep overlooking the easiest source of it — the power already sitting unused inside our own data centers. He’s convinced it’s hiding in plain sight, and he explains it with a kids’ story.

Chapters:

00:00:00 From Boston to Redwood City: an engineer’s “five grand challenges”

00:06:53 Hammerhead AI for non-engineers: the two-toy-car rule

00:09:49 Why data centers run at 30–40% capacity (and why it’s by design)

00:15:41 Plain English: training vs. inferencing, and what a “token” is

00:21:06 Visibility first: how operators stop over-building

00:24:43 Turning stranded power into net-new, flexible workloads

00:28:14 Reliability, community trust, and being a good steward

00:31:56 Resilience by distribution: many sites, one brain (the 60-millisecond point)

00:35:07 From virtual power plants to AI orchestration

00:41:46 Small Is Beautiful vs. The Bitter Lesson: the tension driving the company

00:49:08 The “easy button,” and the one big idea: a utilization crisis, not a power crisis

References and Links:

Rahul Kar (guest) — LinkedIn: linkedin.com/in/rahulkar

Hammerhead AI (guest’s company): hammerheadco.ai

AutoGrid — Rahul Kar’s prior company; pioneered software-defined “virtual power plants” and was later acquired by Schneider Electric 

E. F. Schumacher, “Small Is Beautiful” (1973) — the economics-of-scale idea Rahul cites as a guiding influence

Richard Sutton, “The Bitter Lesson” (2019) — the essay arguing that brute-force compute tends to beat hand-designed methods; Rahul calls Sutton “the father of reinforcement learning” Dean Nelson — named as a longtime champion of the do-more-with-what-you-have mindset; associated with Sun Microsystems

Pacific Telecommunications Council (PTC) — the industry conference where David spoke on a panel Rahul attended

Thank you for visiting our channel, be sure to Like, Comment, and Subscribe!

For more information, go to https://www.QTSDataCenters.com

Learn More about QTS Data Centers on the QTS YouTube Channel: https://www.youtube.com/users/qtsdatacenters

Think you have a great topic for a podcast or would like to schedule an interview with Dave? Send in your requests to podcast@qtsdatacenters.com

Most people think the cloud floats. Saurav Roychoudhury, a Capital University economist, ran the numbers on Ohio’s data centers and found concrete, copper, and a whole lot of jobs. We get into ripple effects, why community trust is the real bottleneck, and whether any of this is zero-sum.

Chapters:

00:00:00 From Calcutta to Capital University: the unlikely data center economist

00:02:31 How the Ohio Chamber’s data center study came to be

00:04:58 “It’s not ethereal”: what the cloud is actually made of

00:05:44 Inside the method: IMPLAN modeling and 30-plus interviews

00:07:53 The real bottleneck — power and water are solvable, trust is harder 00:08:42 “We want it, just not in our backyard”: the clean-water parallel

00:10:58 The mystery box: what’s really running inside a data center

00:19:11 Why the job count lags: a federal code updated only every five years

00:25:09 Ripple effects, and why above-median wages matter

00:35:06 Ohio, the third-largest data center market in the U.S.

00:40:11 Data centers vs. the dot-com boom and the railroads: a GDP comparison

00:47:48 A view from India: 1.4 billion people, less capacity than New Albany

00:50:04 Send an evangelist, not just engineers: how to earn community trust

01:08:00 New Albany’s tax innovation and the Chromebook check

01:24:05 Behind-the-meter power and a stronger grid for everyone

01:28:05 “What are we for?” Why this isn’t a zero-sum game

Links:

Saurav Roychoudhury — LinkedIn: linkedin.com/in/sauravrc

Capital University — Economics: capital.edu/academics/majors-and-minors/economics

The Economic Impact Study of Data Centers in Ohio — by SRC EvalMetrics LLC (the guest’s firm) for the Ohio Chamber of Commerce Research Foundation: ohiochamberfoundation.com — data center study PwC, “Economic, Environmental, and Social Impacts of Data Centers in the United States” — commissioned by the Data Center Coalition (referenced ~00:27:32): datacentercoalition.org/reports-and-publications

IMPLAN — the input-output economic modeling system used in the study (~00:05:44): implan.com

Thank you for visiting our channel, be sure to Like, Comment, and Subscribe! For more information, go to https://www.QTSDataCenters.com

Learn More about QTS Data Centers on the QTS YouTube Channel: https://www.youtube.com/users/qtsdatacenters

Think you have a great topic for a podcast or would like to schedule an interview with Dave? Send in your requests to podcast@qtsdatacenters.com

Join us for a live discussion from Truist Park at the 2026 Atlanta Conn3ct Data Summit where we discuss the very real challenge of educating and attracting the work force of the future.

Thank you for visiting our channel, be sure to Like, Comment, and Subscribe! For more information, go to https://www.QTSDataCenters.com

Learn More about QTS Data Centers on the QTS YouTube Channel: https://www.youtube.com/users/qtsdatacenters

Think you have a great topic for a podcast or would like to schedule an interview with Dave? Send in your requests to podcast@qtsdatacenters.com

Cosme Garcia didn't plan to work in data centers — a project manager got sick, somebody asked him to fill in, and three decades later he's still here, still curious, and still trying to be the least smart person in the room. We talk about wrenches, weather, and what to tell your kid.

Chapters:

00:00 How Cosme Accidentally Ended Up in Data Centers

02:00 Exodus Communications and Learning Power and Cooling the Hard Way

08:30 Why Being the Least Smart Person in the Room Is the Goal

10:30 The Question Cosme Is Asking: When Do We Use AI on Ourselves?

14:00 From a Refrigerator-Sized Gigabyte to a 250-Gigabyte Phone

18:40 Even the Lineman on the Pole Now Depends on a Massive Building

22:00 The Cold War Between IT Teams and Facilities Teams

29:30 Jim Gao, DeepMind, and the Variable Nobody Predicted

35:00 Why Humans Are Bad at Patterns and Casinos Are Good at Them

41:30 Too Many AI Tools, Not Enough Time: Where Do You Even Start?

46:30 Agentic AI: Beyond Prompts, Toward Work That Does Itself

54:00 Trust, IBM, and Picking a Partner When the Market Is Crowded

59:30 Are Our Jobs at Risk? Which Ones, and Why a Wrench Still Wins

1:06:00 QTS Grew 20-Fold; Staffing Didn't. Here's What Happened.

1:14:00 What Will Force the Next Wave: Power, Competition, and One Senior Leader

1:21:00 Advice for the 26-Year-Old: Be Fearful or Be the Person Who Figured It Out

Reference Links: Google / DeepMind data center efficiency work — the project Jim Gao led that applied machine learning to data center cooling optimization. Public reference: https://deepmind.google/discover/blog/deepmind-ai-reduces-google-data-centre-cooling-bill-by-40/

Cosme Garcia on LinkedIn: https://www.linkedin.com/in/cosmegarcia/

Thank you for visiting our channel, be sure to Like, Comment, and Subscribe!

For more information, go to https://www.QTSDataCenters.com

Learn More about QTS Data Centers on the QTS YouTube Channel: https://www.youtube.com/users/qtsdatacenters

Think you have a great topic for a podcast or would like to schedule an interview with Dave? Send in your requests to podcast@q.com