The Intel Xeon E7-8800 v3 Review: The POWER8 Killer?
by Johan De Gelas on May 8, 2015 8:00 AM EST- Posted in
- CPUs
- IT Computing
- Intel
- Xeon
- Haswell
- Enterprise
- server
- Enterprise CPUs
- POWER
- POWER8
The Competitor: IBM's POWER8
As we briefly mentioned in the introduction, among all of the potential competitors for the Xeon E7 line, IBM's OpenPower might be the most potent competitor at this time. So how do IBM's offerings compare to Intel's? IBM POWER 8 is a Brainiac (high IPC) design that also wants to be speed demon (high clock speeds).
The POWER8 core can decode, issue and execute and retire 8 instructions per cycle. That degree of of instruction level parallelism (ILP) can not be extracted out of (most) software. To battle the lack of ILP in software, no less than 8 threads (SMT) are active per core. According to IBM,
- 2-threads delivers about 45% performance more than one
- 4-threads deliver yet another 30% boost
- the last 4-threads deliver about 7%
So in total, the 8-way SMT doubles the performance of this massive core. Let us compare the two chips.
| Xeon E7v3/POWER8 Comparison | ||
| Feature | Intel Haswell-EX Xeon E7 |
IBM POWER8 |
| Process tech. | 22nm FinFET | 22nm SOI |
| Max clock | 2.5-3.6 GHz | 3.5-4.35 GHz |
| Max. core count Max. thread count |
18@2.5 GHz 36 SMT |
12@4.2 GHz 96 SMT |
| Max. sustained IPC | 6 (4) | 8 |
| L1-I / L1-D Cache | 32 KB/32 KB | 32 KB/64 KB |
| L2 Cache | 256 KB SRAM per core | 512 KB SRAM per core |
| L3 Cache | 2.5 MB SRAM per core | 8 MB eDRAM per core |
| L4 Cache | None | 16 MB eDRAM per MBC (64/128 MB total) |
| Memory | 1.5 TB per socket (64 GB per DIMM) |
1-2 TB per socket (64 GB per DIMM) |
| Theoretical Memory Bandwidth | 102 GB/s (independent mode) |
204 GB/s |
| PCIe 3.0 Lanes | 40 Lanes | 32 Lanes |
The POWER8 looks better than Haswell-EX in almost every spec, but the devil is of course in the details. First of all, Intel's L2-cache works at the same clock as the core, IBM's L2-cache runs at a lower clock (2.2 GHz or less, depending on the model). Secondly, the POWER8's L3 eDRAM cache might be much larger, but it is so also a bit slower.
But the main disadvantage of the POWER8 is that all this superscalar wideness and high clockspeed goodness comes with a power price. This slide from Tyan at the latest OpenPOWER conference tell us more.
A 12 core POWER8 is "limited" to 3.1 GHz if you want to stay below the 190W TDP mark. Clockspeeds higher than 4 GHz are only possible with 8-cores and a 250W TDP. This makes us really curious what kind of power dissipation we may expect from the 4.2 GHz 10-core POWER8 inside the expensive E870 Enterprise systems (300W?).
That is not all. Each "Jordan Creek2" memory buffer on the Intel system is limited to about 9W. IBM uses a similar but more complex "Centaur" memory buffer (including a 16 MB cache) which needs more than twice as much energy (16-20W). There are at least four of them per chip, and a high-end chip can have eight. So in total the Intel CPU plus memory buffers have a 201W TDP (165W CPU + 4x9W Jordan Creek 2), while the IBM platform has at best a 270W TDP (190W CPU+ 4x20W MBC).
Facebook
Twitter
RSS
146 Comments
View All Comments
kgardas - Thursday, May 21, 2015 - link
@Kevin G: thanks for the correction about load/store units doing simple integer operations. I agree with your testing that single-threaded POWER8 is not up to the speed of Haswell. In fact my testing shows it's on the same level like POWER7.So with POWER8 doing 6 integer ops in cycle, it's more powerful than SPARC64 X which is doing 4 or than SPARC S3 core which is doing just 2. It also explain well spec rate difference between M10-4 and POWER8 machine. Good! Things start to be more clear now...
patrickjp93 - Saturday, May 16, 2015 - link
No, just no. Intel solved the cluster latency problem long ago with Infiniband revisions. 4 nanoseconds to have a 10-removed node tell the head node something or vice versa, and no one builds hypercube or start topology that's any worse than 10-removed.Brutalizer - Sunday, May 17, 2015 - link
@patrickjp93,If Intel solved the cluster latency problem, then why are not SGI UV2000 and ScaleMP clusters used to run monolithic business software? Question: Why are there no SGI UV2000 top records in SAP?
Answer: Because they can not run monolithich software that branches too much. That is why there are no good x86 benchmarks.
misiu_mp - Tuesday, June 2, 2015 - link
Just to point out, 10ns at the speed of light in vacuum is 3m, and signalling is slower than that because of the fibre medium (glass) limits the sped of light to about 60% of c and on top of that come electronic latencies. So maybe you can get 10ns latency over 1-1.5m maximum. That's not a large cluster.Kevin G - Monday, May 11, 2015 - link
I am not uninformed. I would say that you're being willfully ignorant. In fact, you ignored my previous links about this very topic when I could actually find examples for you. ( For the curious outsider: http://www.anandtech.com/comments/7757/quad-ivy-br... )So again I will cite the US Post Office using SGI machines to run Oracle Times Ten databases:
http://www.intelfreepress.com/news/usps-supercompu...
As for the UV 2000 not being a scale up sever, did you not watch the videos I posted? You can see Linux tools in the videos clearly that indicate that it was produced on a 64 socket system. If that is not a scale up server, why are the LInux tools reporting it as such? If the UV 2000 series isn't good for SAP, then why is HANA being tuned to run on it by both SGI and SAP?
HP's Superdome X shares a strong relationship with the Itanium based Superdome 2 machine: they use the same chipset to scale past 8 sockets. This is because the recent Itaniums and Xeons both use QPI as an interconnect bus. So if the Superdome X is a cluster, then so is its Itanium 2 based offerings using that same chipset. Speaking of, that chipset does go up to 64 sockets and there is the potential to go that far (source: http://www.enterprisetech.com/2014/12/02/hps-itani... ). It won't be a decade if HP already has working chipset that they've shipped in other machines. :)
Speaking of the Superdome X, it is fast and can outrun the SPARC M10-4S at the 16 socket level by a fatctor of 2.38. Even with perfect scaling, the SPARC system would need more than 32 sockets to compete. Oh wait, if we go by your claim above that "you double the number of sockets, you will likely gain 20% or so" then the SPARC system would need to scale to 512 sockets to be competitive with the Superdome X. (Source: http://h20195.www2.hp.com/V2/getpdf.aspx/4AA5-6149... )
And if you're dead set on an >8 socket SAP benchmark using x86 processors, here is one, though a bit dated:
https://www.vmware.com/files/pdf/partners/ibm/ibm-...
68k - Tuesday, May 12, 2015 - link
You know, those >8 socket systems are flying of the shelf faster than anyone can produce them. That is why Intel only got, according to the article, 92-94% of the >4-sockets market... It seem pretty safe to state that >8 socket is an extreme niche market, which is probably why it is hard to find any benchmarks on such systems.The price point of really big scaled-up servers is also an extreme intensive to think very hard about how one can design software to now require a single system. Some problems absolutely need scale-up, as you pointed out, there are such x86 systems available and have been for quite some time.
Anyone know what the ratio between the 2-socket servers vs >4-socket in terms of market share (number of deployed systems) look like?
68k - Tuesday, May 12, 2015 - link
No edit... Pretend that '>' means "greater or equal" in the post above.Arkive - Tuesday, May 12, 2015 - link
You guys are obviously not idiots, just enormously stubborn. Why don't you take the wealth of time you spend fighting on the internet and do something productive instead?kgardas - Wednesday, May 13, 2015 - link
Kevin, I'll not argue with you about SGI UV. It's in fact very nice machine and it looks like it is on part with latency to Sun/Oracle Bixby interconnect. Anyway, what I would like to note is about your Superdome X comparison to SPARC M10-4S. Unfortunately IMHO this is purely CPU benchmark. It's multi-JVM so if you use one JVM per one processor, you pin that JVM to this processor and limit its memory to the size (max) of memory available to the processor, then basically you do have kind of scale-out cluser inside one machine. This is IMHO what they are benchmarking. What it just shows that current SPARC64 is really not up to the performance level of latest Xeon. Pity, but is fact. Anyway, my point is, for memory scalability benchmark you should use something different than multi-jvm bench. I would vote for stream here, although it's still bandwidth oriented still it provides at least some picture: https://www.cs.virginia.edu/stream/top20/Bandwidth... -- no Superdome there and HP submitted some Superdome results in the past. Perhaps it's not memory scalability hero these days?ats - Tuesday, May 12, 2015 - link
So that must be why SAP did this for SGI: https://www.sgi.com/company_info/newsroom/awards/s...You don't normally recognize partners for innovation for your products unless they are filling a need. AKA SGI actually sells their UV300H appliance.
And SAP HANA like ALL DBs can be used both SSI or clustered.
And the SAP SD 2-Tier benchmark is not at all monolithic. The whole 2-tier thing should of been a hint. And SAP SD 3-Tier is also not monolithic.
Scaling for x86 is no easier nor no harder than for any other architecture for large scale coherent systems. If you think it is, its because you don't know jack. I've designed CPUs/Systems that can scale up to 256 CPUs in a coherent image, fyi. Also it should probably be noted that the large scale systems are pretty much never used as monolithic systems, and almost always used via partitioning or VMs.