How to Save $6000 on a 28-core Flagship Intel Xeon: Platinum 8280 vs Gold 6258R
by Dr. Ian Cutress on August 7, 2020 8:00 AM ESTIn February 2020, Intel launched eighteen new Xeon Scalable second generation processors. These mid-cycle additions to Intel’s product portfolio were designed to bolder up Intel’s server offerings on a very popular and very successful platform, adding in extra cores, extra frequency, or more cache than the previous offerings at roughly the same price. The goal of these ‘performance-oriented’ processors was to address customer requests in offering a more palatable performance-per-dollar offering. One of the new CPUs caught our eye: the Xeon Gold 6258R.
Skylake, Cascade Lake, Refresh
Colloquially known as ‘Cascade Lake Refresh’, these processors are the same silicon as the second generation Cascade Lake Xeon Scalable processors that were originally launched in April 2019. In most cases, the Refresh processors focus on both performance and performance-per-dollar metrics, especially given that Intel’s competition in this space were in a very competitive position and focusing on those values. Despite Intel’s data center revenue growing rapidly through 2019 and into 2020, there was a need to effectively replace or add new products into the areas where Intel believed it could keep a strong grasp on the customer base.
In our original announcement for the refresh parts, Intel touted an average performance gain of 36%, and a performance-per-dollar of 42%, although that was pictured as a 1st Gen to 2nd Gen Xeon Scalable jump. For a lot of the eighteen new processors on offer, they either added extra cores, more cache, or more frequency for the same cost as the parts they effectively replace. This usually comes with an increase in power consumption (there’s no escaping the physics), given there was no actual change to the underlying silicon, it simply was a function of binning and product margin.
One of the new parts was the Xeon Gold 6258R, with the R indicating ‘Refresh’. This processor was actually the highest core count refresh part, offering 28 cores at 2.7 GHz base and 4.0 GHz turbo within 205W.
For anyone who follows Intel’s server processor portfolio, those specifications look *very* familiar. Looking through the list, there is one very popular processor that has the exact same specifications: the Xeon Platinum 8280. Here’s the full breakdown:
Intel 2nd Generation Xeon Scalable 28-Core Comparison |
||
Platinum 8280 |
AnandTech | Gold 6258R |
28 Cores / 56 Threads | Cores / Threads | 28 Cores / 56 Threads |
2700 MHz | Base Frequency | 2700 MHz |
4000 MHz | Turbo Frequency | 4000 MHz |
38.5 MB | L3 Cache | 38.5 MB |
3 x 10.4 GT/s | UPI Links | 3 x 10.4 GT/s |
8 | Max Socket Suport | 2 |
6 x DDR4-2933 | DDR4 Support | 6 x DDR4-2933 |
1 TB | DDR4 Capacity | 1 TB |
LGA3647 | Socket | LGA3647 |
205 W | TDP | 205 W |
$10009 | List Price | $3950 |
The Platinum 8280 and the Gold 6258R are identical, almost to a fault. The same cores, the same frequency, the same power, and both support Optane DCPMM. The implementation difference is very subtle: where the 8280 supports 8-way socket deployments, the 6258R only supports 2-way. Intel has separated up the 8200 series and the 6200 series in this sole difference of socket support, which is actually more a firmware difference than anything else.
Oh, and the 6258R has a list price over $6000 cheaper.
Now, the reason why this is important comes down to where the 8280 sits in Intel’s Xeon portfolio. It is, for all intents and purposes, the processor that gets the most attention. It sits at the top of its public processor offerings*, it offers the most cores, and the list price is $10009**. If a non-technical executive is requesting ‘the best’ hardware for deployment, they naturally scroll to the most expensive part and add-to-basket. That processor would be the Xeon Platinum 8280.
However, most servers are single socket and dual socket, which essentially nullifies the ‘extra’ 4-socket and 8-socket capability that the Xeon 8280 offers. In this case, Does the 6258R, with the same specifications on core count, frequency, and power, perform the same as the 8280 but at a fraction of the price?
This is the question I set out to answer with access to both CPUs. Saving $6000 per single socket server, or $12000 in a dual socket configuration, would allow purchasers to focus that investment in other areas, such as memory or storage, or bring down the cost of purchasing quite considerably.
Footnotes
*Intel also offers a Xeon Platinum 8284 which also has the 28 cores that the 8280 does but is at a higher base frequency (same turbo) and 240W TDP. The list price is $15460, a +50% jump. This processor doesn’t seem to always be available everywhere, plus it was also launched months after the 8280.
**List prices from Intel are usually set as the price if someone buys 1000 units, so one would expect the individual cost would be slightly higher. However, major OEM partners and big hyperscalers rarely pay the list price, and the separate pricing is negotiated by contract. Rumors are that the big companies that might need a 100k units or more rarely pay more than 20-50%% of the list price. Exact figures are hard to come by.
81 Comments
View All Comments
benedict - Friday, August 7, 2020 - link
The question is not whether Intel is shooting itself in the foot.It is whether anyone buying Intel is shooting himself in the foot.
In the past no one got fired for buying Intel. Maybe it's time to change that.
ZoZo - Friday, August 7, 2020 - link
It's not just about performance or performance/watt, it's also about platform features and robustness. If you buy Intel you're a bit more certain that things will "just work".For example, I bought a 3rd gen Threadripper and have encountered incompatibilities with Linux KVM-based virtual machines, whether it's the FLR reset bug on the USB controllers when passing them through (fixed late June in Linux kernel, 7 months after TR was released), the fact that a Windows guest doesn't yet support nested AMD virtualization (coming in 2H20), or strange performance behaviors that don't happen on Intel (no resolution in sight). These quirks are bound to exist on Epyc too, but I'll admit that virtualization is probably the most tricky thing you can put the platform through. If you're building a server that just takes a bunch of containers without any virtualization gimmicks, it should be fine.
For workstations, the Intel platform supports RDIMMs and therefore much more RAM, unless you buy from the very few OEMs that sell the Threadripper Pro.
Revv233 - Friday, August 7, 2020 - link
You put that very well about the expectations of not having wierd issues.I always kind of figured it was due to chipset more than CPU but I've experienced that from my old Barton & A64's in spades...
To be fair I once had a P4 Northwood on a VIA chipset and I felt that same pain.
When you are talking mission critical stuff. One bad taste from 20 years ago is going to make you hesitate before you give something another try.
eek2121 - Sunday, August 9, 2020 - link
I agree with him. I run AMD all day long, but at work we tried that with Rome, apparently. Something about our application stack didn’t work well with EPYC. I can only assume it was memory latency, however I am not in that department so I don’t know. What I DO known is that AMD has an amazing product, but the platform isn’t there yet. They either need to clamp down on OEMs or release a first party platform to push the OEMs to release better quality stuff.yeeeeman - Friday, August 7, 2020 - link
I can't say this enough times....I see a lot of comments about AMD being now all of a sudden THE CHOICE.
Hold on for a bit, since there is much more to it than performance or efficiency.
Performance wise, sure, AMD has an advantage, but it is not out of the ordinary. Efficiency wise, the same.
Quality wise, there is no comparison between Intel and AMD platforms. AMD has a lot of work to do on the software and even hardware/firmware front. Many people are not aware of this maybe, but Intel during all these years has invested a lot of effort and money in creating streamlined platforms, with quality software, quality firmware, that it is almost plug and play.
AMD on the other hand, being the underdog and so far away of Intel for so long, nobody focused on their products and they lack badly in optimization and compatibility. Sure, things will change with time, but it will take at least 5 more years of actual work and $$$ from AMD to make it happen.
Same is valid for notebook space. Everyone is crying about how AMD laptops are crippled and how OEMs love Intel. Well, Intel has invested a lot of money and effort in creating design templates and making it as easy as possible for OEMs to create a new laptop.
AMD....well, they just have a good cpu and that is it. There is a lack of field engineers, a lack of streamlined process, a lack of clear BOM and especially a perception from the market that still sees AMD as the cheaper option. Non hardware guys, which is basically 90% of the market cannot deduct from a sticker that 4th gen Ryzen is much better than Buldozer or whatever. AMD needs to invest $$$ into publicity, into OEM partnerships, into creating something similar to project Athena, something to give them the premium feel so that the market perception will change. Otherwise, it will take years before they actually reach majority of market share and by that time Intel could come back.
Anyway, back to the topic of this review, 10k $ is in any case a ridiculous amount of money for an old CPU. 3k $ is even stretching it, so you say it right. If someone wants a top Intel CPU, they should buy this Gold version.
duploxxx - Saturday, August 8, 2020 - link
I can't say this enough. You talk bias. You don't need field engineers. Bom are delivered by oem like Dell hpe Cisco that increase there portfolio for amd on every release. Intel is not investing a lot of money in business for stability all they do is making sure business dies not get fed up with all the CVE bugs they need to patch and deliver in fact the long lasting issues with supplies had them against the wall big time. All intel does is paying money to oem in r&d to keep designing base lines so that they can keep selling the masses. This is done in both WS as Server area. But let's be honest AMD would never be capable to deliver far more. But it's thx to AMD that things like Skylake R exist don't forget that.Smell This - Saturday, August 8, 2020 - link
Now with new glue and Omni-Path v4.0.89.42.000
(Two Dies Are Twice As Nice As One!)
The Cascade Lake Xeon Scalable platform failed last year, and the refresh will fail, again, today.
ProDigit - Saturday, August 8, 2020 - link
Who cares about optimization, when their processors are 25% more efficient, and host between 25-150% more cores?DominionSeraph - Monday, August 10, 2020 - link
This.When the Ryzen 2700 dropped to $150 I jumped on it, figuring I could retire my i7 4790. I put it together and its encoding speed was a very impressive improvement. I couldn't swap out my main immediately as the mobo didn't have enough SATA ports for all my drives so I was using them side-by-side for a while, and I noticed the AMD just lagged and had weird quirks where the 4790 was perfection 24/7/365. I couldn't live with it and gave up on the idea of using it as a replacement, and I eventually sold it off.
I'm still on the i7 4790 as my main even though I now have a 3950X too. AMD is ok for a secondary crunching machine but it's just not suitable for human use.
WizardMerlin - Wednesday, August 26, 2020 - link
Having been using my 3900X for many months now for 16 hours a day for a varied workload, I'd disagree.