Weight. If you load a 42u rack up with 30lb servers you’re at 1280lbs spread out over about 4sqft, which is over the floor loading limit for most buildings. It’s much cheaper to support the weight in a wider building compared to a taller one.
That being said I’ve been in many data centers in the middle of a giant office towers, but they have lower weight limits generally.
20 years ago I worked on the top floor of a 5 story office building. We wanted to build out a server room with a pretty hefty UPS for backup. The amount of steel reinforcement we had to install in the ceiling of the 4th floor was pretty insane…
This is armchair nonsense. And a few down votes don’t change this.
Slab reinforcement in skyscrapers can literally hold fucking skyscrapers.
If you want to call this a limit, the limit is the expense. Not the weight. That’s absurd. We’re so far past that.
Edit: proof bc this guy is overconfidently incorrect. Here. Here are plenty of skyscraper data centers that exist. Weight did not limit them. The cost of available land dictated that the cost of slab reinforcement was cheaper to build upwards than outwards. It is rare. Cost is the limiting factor.
I’ve deployed substantial quantities of gear in 9 datacenters across 4 countries in my career. I’ve gotten a panicked call from Bell Canada when they realized our deployment density in an older facility, then had to work with them to provide weights of all of our cabinets. Sure though, all armchair nonsense. What’s your background?
I’m just going to repeat myself. Ildeploying a day center isn’t building one. If you’d architected a data center I’d give a fuck about what you’re saying
Again:
Cost is the limiting factor. Not weight. You are provably incorrect. You can ignore proof all you want. It doesn’t make you right.
You started off with an ad hominem attack calling it armchair nonsense, that the weight argument has no merit. I’m pointing out i have actual experience in this area. If you hadn’t been an asshole with your initial reply I wouldn’t have bothered replying, instead here we are.
Everything is a money problem when you get down to it far enough. Why don’t we have mars colonies? Money. Why don’t cars fly? Money. Why doesn’t everyone live in super tall towers that touch the atmosphere? Money. Sure let’s just ignore all the engineering considerations and reduce it down to the absolute basic explanation of “money” so that nobody in this thread will learn anything.
Why don’t we have super tall datacenters? It’s not worth the money to sustain that level of weight in a new tower, and definitely not worth it to overhaul an existing tower.
It’s pointless to call out money as the limit, that’s completely obvious.
I did not make an ad hominem attack. Tf are you talking about?
We LITERALLY do have tall (not super tall, but over a dozen stories) data centers, you overconfidently and provably incorrect human.
We build fucking residential mega towers with swimming pools larger than the average lake on the roof. It is cost. It is not a hard limit. It is cost. You overconfident idiot.
They’re (tall data centers) in Manhattan, Tokyo, etc. Where the aforementioned cost of that weight does not surpass the cost of la d sprawl.
They don’t get ‘super’ tall bc heat goes up. Not because of weight. (Handling that heat is also cost, it is less exponential).
But the reason for the expense is largely the weight.
Yes we can at great expense support massive weights. But even in skyscrapers, you aren’t expecting to just cram every floor with equipment that weighs over a ton and supported by less than a square meter of floor.
It’s not just armchair engineering, i work in the industry and commonly you have racks preferring the ground floor and weight restrictions going up and even marked paths that the racks need to stay on when on upper floors due to limitations of the reinforcements.
Skyscrapers are largely impractical structures done for the sake of showing off, with any value based on keeping people close to each other. No one builds a skyscraper by itself miles from anything else. This is where they build the datacenters because they don’t need proximity.
Weight. If you load a 42u rack up with 30lb servers you’re at 1280lbs spread out over about 4sqft, which is over the floor loading limit for most buildings. It’s much cheaper to support the weight in a wider building compared to a taller one.
That being said I’ve been in many data centers in the middle of a giant office towers, but they have lower weight limits generally.
20 years ago I worked on the top floor of a 5 story office building. We wanted to build out a server room with a pretty hefty UPS for backup. The amount of steel reinforcement we had to install in the ceiling of the 4th floor was pretty insane…
This is armchair nonsense. And a few down votes don’t change this.
Slab reinforcement in skyscrapers can literally hold fucking skyscrapers.
If you want to call this a limit, the limit is the expense. Not the weight. That’s absurd. We’re so far past that.
Edit: proof bc this guy is overconfidently incorrect. Here. Here are plenty of skyscraper data centers that exist. Weight did not limit them. The cost of available land dictated that the cost of slab reinforcement was cheaper to build upwards than outwards. It is rare. Cost is the limiting factor.
NOT WEIGHT.
you’re so
provably
Wrong
Like jfc bro. Here:
Sabey NYC (375 Pearl)
Digital Realty (60 Hudson)
NYI (60 Hudson)
CoreSite (32 Ave of the Americas)
Digital Realty (111 8th Ave)
Equinix TY11 (Tokyo Ariake)
AT TOKYO (Chuo Center)
Telehouse (Tokyo Otemachi)
Equinix TY4 (Tokyo Otemachi)
Now fuck off with your provably incorrect armchair bullshit.
The limit is cost.
Not weight.
I’ve deployed substantial quantities of gear in 9 datacenters across 4 countries in my career. I’ve gotten a panicked call from Bell Canada when they realized our deployment density in an older facility, then had to work with them to provide weights of all of our cabinets. Sure though, all armchair nonsense. What’s your background?
30 seconds searching will back me up. https://www.digitalrealty.com/resources/articles/what-floor-loading-capacity-do-dlr-data-centers-have
I’m just going to repeat myself. Ildeploying a day center isn’t building one. If you’d architected a data center I’d give a fuck about what you’re saying
Again:
Cost is the limiting factor. Not weight. You are provably incorrect. You can ignore proof all you want. It doesn’t make you right.
You started off with an ad hominem attack calling it armchair nonsense, that the weight argument has no merit. I’m pointing out i have actual experience in this area. If you hadn’t been an asshole with your initial reply I wouldn’t have bothered replying, instead here we are.
Everything is a money problem when you get down to it far enough. Why don’t we have mars colonies? Money. Why don’t cars fly? Money. Why doesn’t everyone live in super tall towers that touch the atmosphere? Money. Sure let’s just ignore all the engineering considerations and reduce it down to the absolute basic explanation of “money” so that nobody in this thread will learn anything.
Why don’t we have super tall datacenters? It’s not worth the money to sustain that level of weight in a new tower, and definitely not worth it to overhaul an existing tower.
It’s pointless to call out money as the limit, that’s completely obvious.
Anyways I’m over this thread, byeeeee.
I did not make an ad hominem attack. Tf are you talking about?
We LITERALLY do have tall (not super tall, but over a dozen stories) data centers, you overconfidently and provably incorrect human.
We build fucking residential mega towers with swimming pools larger than the average lake on the roof. It is cost. It is not a hard limit. It is cost. You overconfident idiot.
They’re (tall data centers) in Manhattan, Tokyo, etc. Where the aforementioned cost of that weight does not surpass the cost of la d sprawl.
They don’t get ‘super’ tall bc heat goes up. Not because of weight. (Handling that heat is also cost, it is less exponential).
Tagged as overconfidently incorrect…
Edit: you’re so
provably
Wrong
Like jfc bro. Here:
Sabey NYC (375 Pearl)
Digital Realty (60 Hudson)
NYI (60 Hudson)
CoreSite (32 Ave of the Americas)
Digital Realty (111 8th Ave)
Equinix TY11 (Tokyo Ariake)
AT TOKYO (Chuo Center)
Telehouse (Tokyo Otemachi)
Equinix TY4 (Tokyo Otemachi)
Now fuck off with your provably incorrect armchair bullshit.
Again:
Cost is the limiting factor. Not weight. You are provably incorrect. You can ignore proof all you want. It doesn’t make you right.
Byeeee
deleted by creator
But the reason for the expense is largely the weight.
Yes we can at great expense support massive weights. But even in skyscrapers, you aren’t expecting to just cram every floor with equipment that weighs over a ton and supported by less than a square meter of floor.
It’s not just armchair engineering, i work in the industry and commonly you have racks preferring the ground floor and weight restrictions going up and even marked paths that the racks need to stay on when on upper floors due to limitations of the reinforcements.
Skyscrapers are largely impractical structures done for the sake of showing off, with any value based on keeping people close to each other. No one builds a skyscraper by itself miles from anything else. This is where they build the datacenters because they don’t need proximity.
Weight is not the limiting factor.
Money is.
Proof:
We do have tall (not super tall, but over a dozen stories) data centers, you overconfidently and provably incorrect human.
They’re in Manhattan, Tokyo, etc. Where the aforementioned cost of that weight does not surpass the cost of la d sprawl.
They don’t get ‘super’ tall bc heat goes up. Not because of weight. (Handling that heat is also cost, it is less exponential).
Tagged as overconfidently incorrect…
Edit: you’re so
provably
Wrong
Like jfc bro. Here:
Sabey NYC (375 Pearl)
Digital Realty (60 Hudson)
NYI (60 Hudson)
CoreSite (32 Ave of the Americas)
Digital Realty (111 8th Ave)
Equinix TY11 (Tokyo Ariake)
AT TOKYO (Chuo Center)
Telehouse (Tokyo Otemachi)
Equinix TY4 (Tokyo Otemachi)
Peculiar down votes to this post. Almost like brigading.