News for hackers. Stuff that matters.

A comment on Intel Brings a Big Fork to a Server CPU Knife Fight. Parent comment

Re:Intel Brings a Big Fork to a Server CPU Knife Fight by londons_explore on Wednesday June 05, 2024 @05:41AM (#40583074)
Pretty sure "one VCPU" is scaled to the core performance. So if they have a powerful core that has 1.5x the instruction throughput, it gets sold to users as 1.5 VCPU's.
[ Parent | on HN ]
Re:Intel Brings a Big Fork to a Server CPU Knife Fight by paol on Wednesday June 05, 2024 @06:11AM (#40583258)
Not in AWS. 1 VCPU = 1 thread (so half a core in architectures with 2-way SMT, which is most of them).
[ Parent | on HN ]
Re:Intel Brings a Big Fork to a Server CPU Knife Fight by yencabulator on Wednesday June 05, 2024 @01:20PM (#40587707)
It is, except when it isn't. T2 etc are burstable and each vCPU seems to be guaranteed only 10% of a CPU thread(/core), even while it's called a "vCPU".
[ Parent | on HN ]
Re:Intel Brings a Big Fork to a Server CPU Knife Fight by rbanffy on Wednesday June 05, 2024 @04:23PM (#40590096)
So, in AWS (and most others), a vCPU is something between nothing and one single thread, depending on how much you pay for it.

I wonder why IBM doesn’t do more with POWER10 in their cloud, because each core can go up to, IIRC, 8 threads.

Same for Oracle, BTW, but they are no longer investing on SPARC, just milking until it dries up completely.

[ Parent | on HN ]
Re:Intel Brings a Big Fork to a Server CPU Knife Fight by MaKey on Wednesday June 05, 2024 @05:52AM (#40583144)
At least with Azure that's not the case. Depending on the VM family one vCPU is one core or one thread.
[ Parent | on HN ]