The question of "Does it scale" is just overlooked in this case. The answer is always "just throw more expensive, powerful hardware at it". I always get curious about the resource utilization of such monster server racks.
Something was off about the text in this article. I don't want to allege that it was AI, but it felt AI.
Opening sentence:
At HPE Discover, the company showed some neat hardware. We have hundreds of photos. Luckily, at the end of Wednesday, the show floor was fairly empty, and we grabbed some really neat photos.
Are you telling me a human being wrote that and an editor approved it?
Later on:
Four of these nodes looked different. This is crazy.
If it were the live captions of someone talking about the product transcribed with no further effort, that might make sense, but this tone and irreverence is certainly offputting to me.
As I recall the local cluster at the uni I went to was like 3000 cores, some 20 years ago. Each night the entire cluster was reserved for weather predictions.
The one that replaced it in 2012 was 10000 cores[1]. Pretty wild number, this new rack.
Does one still use MPI for programming these or are there better ways now?
No more "Does it Scale" question. What ever web scaling issues we have, throw in a few rack and call it a day. :)
And these are Real CPU Core, i.e there are 163,840 vCPU per Rack.
Oh this is fun!
The question of "Does it scale" is just overlooked in this case. The answer is always "just throw more expensive, powerful hardware at it". I always get curious about the resource utilization of such monster server racks.
Something was off about the text in this article. I don't want to allege that it was AI, but it felt AI.
Opening sentence:
At HPE Discover, the company showed some neat hardware. We have hundreds of photos. Luckily, at the end of Wednesday, the show floor was fairly empty, and we grabbed some really neat photos.
Are you telling me a human being wrote that and an editor approved it?
Later on:
Four of these nodes looked different. This is crazy.
If it were the live captions of someone talking about the product transcribed with no further effort, that might make sense, but this tone and irreverence is certainly offputting to me.
Imagine a Beowulf cluster of these!
As I recall the local cluster at the uni I went to was like 3000 cores, some 20 years ago. Each night the entire cluster was reserved for weather predictions.
The one that replaced it in 2012 was 10000 cores[1]. Pretty wild number, this new rack.
Does one still use MPI for programming these or are there better ways now?
[1]: https://no.wikipedia.org/wiki/Abel_(superdatamaskin)