Microsoft’s new x86 DataCenter class machines running Windows (2018)

https://www.anandtech.com/show/13522/896-xeon-cores-in-one-pc-microsofts-new-x86-datacenter-class-machines-running-windows

by rbanffy • 6 years ago

22 28 6 years ago

anandtech.com

fortran77• 6 years ago

Why does HN sometimes truncate leading words/numbers in submissions? The actual title is "869 Xeon Cores...."

wmf2• 6 years ago

Those are common clickbait patterns.

derision• 6 years ago

If it's a factual number is it really clickbait?

thoraway10103• 6 years ago

No kidding. Despite the HN posters saying it's "factual" this is likely just 4 socket systems tied together with an interconnect.

If intel is really going to market these type of systems, they need to be a LOT LOT more transparent on interconnect latency / memory management issues etc.

These headlines are almost always clickbait when you try to run high contention workloads on these supposedly massive systems.

What is the name of the interconnect approach being used here? That's what I want to know.

sbierwagen3• 6 years ago

The article states that it’s actually four 8 socket machines connected over pcie. Not what I would call “one PC”

guenthert• 6 years ago

Well, it's clearly not for Personal Computing, nor will anyone run DOS on it. It's a Single System Image (SSI) system. Some DB which can't be (effectively) clustered and other software which doesn't scale out well (or hasn't been rewritten yet to make use of distributed computing) demands such.

saagarjha• 6 years ago

Numbers at the start of titles are stripped, presumably to catch titles such as "10 great tips to improve your business".

thoraway10103• 6 years ago

And in this case it was a good thing. These are just normal machines (4 socket?) tied together with an interconnect.

So the rule works surprisingly well!

johnklos• 6 years ago

Like lipstick on a pig.

Koshkin• 6 years ago

Thank you for the mental image of a compute stick running lisp.

aloknnikhil2• 6 years ago

This is from 2018. Can we add that to the title?

dang• 6 years ago

Good catch. Added. Thanks!

sradman• 6 years ago

Back in the days pre-Nehalem, the big eight socket 32-core servers were all the rage, but today not so much, and unless a company is willing to spend $250k+ (before support contracts or DRAM/NAND) on a single 8-socket system, it’s reserved for the big players in town. Today, those are the cloud providers.

SAP HANA is a major driver for single servers with the maximum number of cores and maximum memory per core that Intel supports. This is true for on-premise as well as Azure [1] and AWS [2] IaaS.

I think of it as commodity scale-up and the superior price-per-performance applies to any hybrid in-memory column store though HANA seems to be the only offering that exploits this approach. You can store a lot of data in 24 TB (or more) of in-memory compressed bitmap indexes.

[1] https://docs.microsoft.com/en-us/azure/virtual-machines/work...

[2] https://aws.amazon.com/sap/solutions/saphana/

pgtan• 6 years ago

SAP offers similar technology on their Enterprise Cloud using POWER9 interconnected nodes:

https://insidesap.com.au/sap-hec-to-run-on-ibm-power-system-...

https://www.youtube.com/watch?v=djtqJ4P3OnU

fomine3• 6 years ago

Fun quote:

From what we have heard from both OEMs and Intel, the vast majority of 8-socket servers today are sold in China. The most common reason for this that we hear is not because of the scale-up benefits. Instead, it is because of the numerology of the 8. It just so happens that 8 in China is a lucky number, much like 7 is considered by many in the US to be a lucky number.

https://www.servethehome.com/why-all-servers-are-not-4-socke...

blattimwind• 6 years ago

The late 1990s called and want their Numalink back.

p_l• 6 years ago

Some of the successful large-socket-count servers, specifically SGI Ultraviolet series (now sold by HPE AFAIK) actually use NUMAlink - a much updated version, but the basic remain the same as with MIPS and Itanium systems.

aloknnikhil2• 6 years ago

From one of the comments on the article

I think Intel / Microsoft have found a way to interconnects scalable and this comes interesting thought on value of have more cores on cpu. Especially if system was designed to be pluggable and if one of cpu's failed than it does not bring entire system down.

I think this is an interesting point. If the QPI links on the Xeon Platinum is restricted to 3, I wonder what the 32 socket mesh looks like and what implications that has on availability of the system as a whole if one CPU was to fail. Also, unless these are fully NUMA aware workloads or each socket was dedicated to a VM, the latency has to be terrible for memory access especially if the interconnect is multiple hops away. Clearly it's not going to be a fully connected mesh?

freeqaz• 6 years ago

This starts to blur the line in my head between the low-level system concepts that your compiler/OS has to think about and the concepts you have to think about for distributes systems.

With the hardware systems and interconnects today, we've drawn a line in the sand based on what we've seen work and fail. Something like Redis is fairly well understood for many architectures. But when you have crazy NUMA architectures and different memory models like Intel Optane, what changes? Where do you draw the "new" line for when to use Redis vs keep everything "local" to one box?

Will we see a local Kubernetes extension (or new software) that makes building for "massive boxes" easier?

Kind of a cool idea.

Microsoft’s new x86 DataCenter class machines running Windows (2018)

Related Stories