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
Santoval - Friday, August 7, 2020 - link
Virtualization is tricky for Ryzen, but it is not tricky for Threadripper and it is *certainly* not tricky for Epyc. AMD need to fix their shit in order to successfully compete in this market because it's an utter shame if workstation and server customers prefer Intel due to its "robustness" alone.Santoval - Friday, August 7, 2020 - link
p.s. Though, come to think about it, there is nothing "robust" about Intel's multiple security flaws. These matter to server buyers because if things go sideways due to these flaws they are liable to lawsuits by their customers and they can also suffer damage to the reputation and trustworthiness of their company.Santoval - Friday, August 7, 2020 - link
p.s. By the way, the "strange performance behaviors that don't happen on Intel (no resolution in sight)" bit was way too vague and arbitrary. As stated it means nothing, so could you please clarify?ZoZo - Friday, August 7, 2020 - link
I did some benchmarks under Linux, and some of them (a small minority) showed some drastic performance drops when run in virtual machines under KVM (kernel 5.7), as much as 80% in one case. That did not happen when I tested with a Xeon W-2295 before. I spent hours trying to find what was wrong, tinkering with many variations of settings for the VMs. Turns out switching from KVM to ESXi 7.0 kind of fixed the problem. So most probably not a problem with the silicon, but it still showed me that software support can be shaky in some areas.eek2121 - Friday, August 7, 2020 - link
Honestly, I tend to look past a lot of the defenses for Intel, but claiming perf/watt doesn't matter? At least I know you aren't in IT. I deal with a very large amount of companies where the difference of a single watt can determine whether dozens of people have jobs or not. Why? because a single server may not matter much, but thousands? they matter a lot .ZoZo - Saturday, August 8, 2020 - link
I'm not trying to defend Intel, what is it with all you binary-thinking people. I just bought AMD for myself and after some struggle am finally happy with it. I'm only telling what I suspect could be a reason to still pick Intel today, after what I've experienced with an AMD platform.Where did I say that performance/watt didn't matter? Please, point it out.
I said it's not just about that, meaning that it's not the only thing that might need to be considered.
It's getting annoying to have to explain to compensate for poor reading skills.
WaltC - Friday, August 7, 2020 - link
Yes, and how do we know it hasn't changed already?...;) I think it has. Hallelujah, it's about time! This article reminds me of the Intel monopoly Halcyon days when Intel had no high-end x86 competition--the years after the Operon/A64 peak but prior to the AMD Zen debut with Zen 2 a year ago. Let's hope that Intel would iron out most problems in its architectures after essentially rehashing them for many years...;) Milking the cow until the cow ran dry...! Time for something new, Intel..chop, chop...!yankeeDDL - Friday, August 7, 2020 - link
6000 premium for more socket support, seems precisely one of those tricks played by a market monopolist that I am glad to see ending.Spunjji - Friday, August 7, 2020 - link
Indeed. One wonders what the point of setting the price that much higher really is if, in fact, none of their customers ever actually pay that price. Perhaps it's merely to make giving "discounts" that much easier, and/or to provide an opportunity to milk the wealthiest customers.MrVibrato - Friday, August 7, 2020 - link
You are probably right. It is very likely a sales tactic. Those CPUs are not sold in retail (at least not in quantities), but as part of enterprise/complex sales, where contracts (including prices) are negotiated. It is good negotation tactic for a seller to start with a high(er) price. Everyone who ever visited a bazar and haggled with a merchant knows how it goes.And despite having no evidence or experience of Intels approach to sales and negotation, i doubt that anyone who buys more than a handful of those puppies would pay anything close to the listed price...