Our man pops the hood on Intel's v4 engine: Broadwell Xeons

Recently, I reviewed Supermicro's Microblade system. One of the goals of this review was to compare the new Intel v4 (Broadwell) Xeons to their predecessor v3 (Haswell) Xeons. This was not as easy as it should have been. My front line tool for benchmarking CPUs is Prime 95. Supermicro provided me with 2x Intel Xeon E5 2695 v4 …

[Bronek Kozicki]

Thanks for that

... also for reference to The Platform, the tables there are most useful. One more thing worth noting is (re)introduction of TSX-NI, which in theory should also kick without any program modifications, for optimistic locking of synchronized sections. I wonder how that works in practice, any ideas?

[Fenton]

Core Count

Even if performance per core per Ghz is the same,

Going from v1 to v4 will give you a massive improvement in core count per socket.

Which will dramatically reduce the number of physical servers required in a virtualized environment.

[GrumpenKraut]

Just a short thanks for the article.

Thanks Trevor! --------------->

[batfastad]

Ouch

Imagine the Oracle licensing cost for one of these!

[Anonymous Coward]

Re: Ouch

Microsoft too, now.

[Disgruntled of TW]

Re: Ouch

Yep - 8 core sockets is the magic number, which if you go higher will cost you more in Server licence.

[Paul Smith]

Re: Ouch

I would see that as encouragement to explore non-proprietary options.