Open Compute Project seeks standard for concrete, with help from AWS, Google, Meta, and Microsoft

The Open Compute Project, the org best known for offering designs for hyperscale hardware, has rounded up AWS, Google, Meta, and Microsoft to help it test concrete. In a blog post on Tuesday, the Project cited a 2023 open letter from carbon reduction advocacy org iMasons' Climate Accord that points out that concrete is a …

[Pascal Monett]

"a thoroughly worthy exercise"

Agreed. Anything we can do to lower our carbon emissions is a worthy goal, in my view.

As far as concrete is concerned, until we use it, we won't be able to be sure how it behaves. So every additional test is a Good Thing^TM.

[heyrick]

Re: "a thoroughly worthy exercise"

Just need to be a little bit wary of the next best thing, we don't need another infrastructure mess like reinforced autoclaved aerated concrete.

[Anonymous Coward]

Re: "a thoroughly worthy exercise"

Datacentre lifetimes are measured only in a few decades, because obsolesence of the hardware. It probably won't be the datacentre operator that sees the effect of whatever concrete spec it opts for, but rather, the future owner.

Corroding rebar is the main failing of most reinforced concrete types; with lifetimes somewhere around the 50 to 80 year mark. Lot of motorways and 60's brutalist architecture approaching it's last legs...

The 2000 year old concrete of the Pantheon in Rome is, by comparison, much more long-lived and a damn sight better looking.

[Dunstan Vavasour]

Testing Failure Modes

Always my beef with the new-improved...

I'm not concerned with how it works when new, I'm concerned with all of the failure modes over time. BUT, if the concrete coming out is chemically the same, or similar, to the stuff we get from big carbon producing kilns, I'm all for it.

And yet, I wonder if humanity has leapt into overconsumption of concrete, as it has with plastics, and the better solution is to stop using the stuff.

[Daniel Nebdal]

Re: Testing Failure Modes

One approach is to just do CO2 capture on top of the normal process, and then pump the liquid CO2 down into old oil wells (ref https://www.nationalgeographic.com/premium/article/remove-carbon-emissions ). More expensive than just venting it, but at least the product should be indistinguishable?

[Anonymous Coward]

Call me a professional sourpuss

This Open Compute Project is really a teh zuck hobby project, as was clear in how they started: "Look! We're setting industry standards!" by, instead of using the ubiquitous 19" racks, they went for... 21"!!!!1! Conveniently forgetting that the 'merkin telco people that gave us the 19" also have a 23" on offer. Making adding the 21" a political move, or ineptitude, or both. Moreover, if you're going to reinvent the wheel like that, and be all open and interoperable and shit, then it behooves you to look a little beyond your own little rack here and notice that the rest of the world has moved on to metric. So new standards really need to have that, too, yes, TYVM.

With that in mind I can't help but grouch. "This Open Compute Project making up 'standards' for concrete now? In rods to the hogshead, what?"

[Roland6]

Re: Call me a professional sourpuss

I’m a fan of modern 48.26 cm racking (ie. The variant that uses square punched holes and cage bolts), although it is irritating where some US components (eg. APC Baxter rails) are pre tapped and require UNC or UNF machine screws rather than the more readily available ISO versions…

For those fond of nice round number specifications, I suggest familiarising yourself with how Stevenson having started out with 5 foot ended up with 4 foot 8.5 inches. (bTW it has nothing to do with Roman chariots).

[Sceptic Tank]

Broke their balls

The floor of the Eindhoven Airport parking was stuffed full of plastic balls. See how well that worked, huh? Not much CO₂ savings in pouring new-age concrete if you have to pour again later with the old stuff.

[heyrick]

Re: Broke their balls

Here's a slightly longer version of the embedded video in English. https://youtu.be/jSmRELbmdV4

Worth noting that the problem in this case wasn't the use of the plastic balls, it was rotating the floor panels 90 degrees to span a larger area without paying attention to the consequences of doing so...

[Anonymous Coward]

OCP concrete strength tests

If it can support ED-209 then it passes.

Just make sure there aren't any stairs.