Intel Skylake Counters AMD Epyc

Release time:2017-07-12
source:Rick Merritt

Intel announced that its Skylake server processors sport an average 65% performance boost over its prior Broadwell chips. Top-end versions of the new Xeon Scalable family nudge ahead of rival AMD’s recently released Epyc CPUs in performance but don’t pack as much I/O.

The results suggest that Intel will have no problem maintaining its dominance in the lucrative data center. Nevertheless, AMD’s Epyc and a rising tide of ARM-based server chips from Qualcomm and others are expected to find significant footholds in the broad and diverse cloud computing sector.

Platinum 8180 and 8160 versions of Skylake edged AMD’s Epyc 7601 by 2% to 28% in performance and by 12% to 22% in performance/watt on the Specint_rate2006 benchmark. The results could be skewed by Intel’s tendency to use optimized compilers for its benchmarks compared to standard ones that AMD uses.

The high-end 8100 series packs 28 cores running at up to 3.6 GHz with up to 48 PCIe 3.0 lanes and six channels of DDR4-2666 memory. By contrast, AMD’s high-end Epyc packs up to 32 cores and all nine of its family support 128 PCI Express 3.0 lanes and eight DDR4-2666 channels.

Intel showed tests with two dozen companies, each with different workloads. Results ranged from Skylake beating Broadwell chips by 1.4x for Ansys manufacturing software to 2.2x for apps using Skylake along with Intel’s proprietary Optane solid-state memory drives.

The results are “impressive … the increase over Broadwell is much better than they have had in typical generations, especially when you consider [that] these are both 14-nm parts,” said Nathan Brookwood, principal of market watcher Insight64 (Saratoga, California).

Skylake edges past AMD's Epyc, but Intel's optimized compilers may skew results in its favor. (Images: Intel)
Skylake edges past AMD's Epyc, but Intel's optimized compilers may skew results in its favor. (Images: Intel)

“AMD got 25% of the server market when it had a vastly superior product with Opteron, but I don’t think Epyc is vastly better than Skylake,” said Brookwood.

Last month, AMD showed a range of benchmarks for Epyc that averaged around 45% more performance than Broadwell. However, the server sector includes a wide range of markets and requirements, many where Intel will have an edge and a few where AMD may score hits.

For example, AMD hopes to use its advantage in PCIe and DDR4 to replace dual-socket Broadwell with single-socket Epyc servers. However, Skylake’s new AVX-512 vector processing extensions far outstrip Epyc’s abilities in floating-point intensive jobs.

Architecturally, Skylake uses a single processor die with a separate I/O chip. Epyc packs four die in a package including I/O, giving AMD greater flexibility and lower cost at the expense of latency in some operations.

Intel has already shipped more than 500,000 of the chips, which are already running in data centers at Alibaba, Amazon, AT&T, and Google. They are in use at more than 30 customers, including a system in Barcelona ranked as the world’s 13th fastest supercomputer.

Online messageinquiry

AMD President and CEO Lisa Su took to the stage at CES to give what some analysts suggested was the most important appearance by AMD at a tradeshow since a keynote at the now-defunct Comdex by then-CEO Hector Ruiz in 2002. She did not disappoint.Su announced that the Radeon VII GPU would be available starting next month and that its second-generation Ryzen mobile processors would begin showing up in ultrathin and gaming laptops by the end of the first quarter. Su also teased AMD's third-generation Ryzen desktop processor and its second-generation EPYC server processor, both of which are implemented in 7nm and will be available starting later this year.Analysts were enthusiastic about the announcements made by Su Wednesday in the first CES keynote by AMD since 2009, though at least some of the news had been revealed previously. Last November at an event in San Francisco, AMD first announced the 7nm Radeon 7 GPU and the 7nm server processor, which is codenamed "Rome" and based on AMD's Zen 2 x86 microarchitecture. The second-generation Ryzen mobile processors were announced by AMD on Sunday, prior to the start of CES.But Su provided more details Wednesday on all three chips."I think this was a landmark event for AMD," said Kevin Krewell, a principal analyst at Tirias Research. "It was quite dramatic for AMD to usurp the leadership role from Intel."Indeed, with 7nm silicon available next month, AMD will beat both Intel and Nvidia to the most advanced technology node. And by returning to the CES keynote stage, AMD also upstaged its rival, which lost its most coveted CES keynote spot this year to LG Electronics. (Intel, which has yet to replace CEO Brian Krzanich after he abruptly resigned in June, instead held a press conference event Monday).AMD President and CEO Lisa Su shows a 7nm Radeon VII graphics card. (Source: CES)Radeon VII, built on AMD's second-generation Vega architecture, boasts twice the memory and memory bandwidth compared to AMD's current top-of-the-line Radeon. It also improves gaming performance by up to 29% and offers up to 36% better content creation performance compared to the incumbent, AMD said.The device features 60 compute units (or 3,840 stream processors) running at up to 1.8GHz. It also has 16GB of HBM2 memory and 1 TB/s memory bandwidth."We really believe at the high end enthusiast market, you need great hardware," said Su, who devoted most of her 90-minute keynote to the gaming sector.She stressed that gaming requires cutting edge technology and brings people closer together. "Gamers love great technology and they provide us great feedback," Su said.The second-generation mobile Ryzen processors, AMD's Ryzen 3000 Series, are designed for both ultrathin and gaming laptops. The initial 3000 Series chips feature two or four cores with eight or four threads and boost performance of up to 4.0GHz. They will be available in laptops starting later in the first quarter and throughout the rest of the year from the likes of Acer, ASUS, Dell, HP, Huawei, Lenovo and Samsung.Perhaps most notable about the Ryzen 3000 Series is that the devices enable up to 12 hours of general productivity and 10 hours of video playback battery life, according to AMD."If it can actually provide 12 hours of 'productivity battery life" — not fake MobileMark or video playback — this would be big," Moorhead said. "It would be longer than any other notebook. I am looking forward to the reviews and testing myself."Details on the third-generation Ryzen desktop processors remain sketchy. Su showed an early version of the 7nm device going head-to-head with a top-of-the-line Intel Core i9 processor. The third-generation Ryzen desktop processors are based on AMD's Zen 2 microarchitecture and are expected to be the first PC processors to support PCIe 4.0 connectivity. Su said they would be introduced in mid-2019."As a desktop processor, it will absolutely set the bar" in performance, features, and power efficiency, Su said.Su also showed the Rome 7nm datacenter CPU, which is also based on the Zen 2 microarchitecture. She showed a demonstration pitting a pre-production EPYC Rome processor against two high-end Intel Xeon Platinum 8180 processors, with the EPYC processor showing about 15% greater performance.Among other things, Su said the second-generation EPYC offers four times the floating point performance and twice the computing performance per socket compared to its predecessor. "I will go so far as to say we believe it is the best server processor the world has ever seen," Su said."AMD has achieved phenomenal performance with Zen 2, something that we haven't seen in the x86 architecture for the past decade," said Jim McGregor, principal analyst at Tirias Research. "We finally have competition in PCs and servers from top to bottom. This bodes well for the future in products from both AMD and Intel."Speaking generally about the importance of high-performance computing at the beginning of the keynote, Su said the dramatic increases in the number of connected devices and data generated by them demand dramatic innovation to keep pace, particularly with Moore's Law scaling slowing to a comparative crawl."We are absolutely determined to drive high-performance computing innovation at the bleeding edge," Su said. "This is our job one."
2019-01-10 00:00 reading:512
  • Week of hot material
  • Material in short supply seckilling
model brand Quote
TPIC6C595DR Texas Instruments
TXB0108PWR Texas Instruments
TPS61021ADSGR Texas Instruments
TL431ACLPR Texas Instruments
PCA9306DCUR Texas Instruments
TPS5430DDAR Texas Instruments
model brand To snap up
TPS61256YFFR Texas Instruments
ULQ2003AQDRQ1 Texas Instruments
TPS61021ADSGR Texas Instruments
TPS63050YFFR Texas Instruments
TPS5430DDAR Texas Instruments
TXS0104EPWR Texas Instruments
Hot labels
Information leaderboard
  • Week of ranking
  • Month ranking
About us

Qr code of ameya360 official account

Identify TWO-DIMENSIONAL code, you can pay attention to

AMEYA360 mall ( was launched in 2011. Now there are more than 3,500 high-quality suppliers, including 6 million product model data, and more than 1 million component stocks for purchase. Products cover MCU+ memory + power chip +IGBT+MOS tube + op amp + RF Bluetooth + sensor + resistor capacitance inductor + connector and other fields. main business of platform covers spot sales of electronic components, BOM distribution and product supporting materials, providing one-stop purchasing and sales services for our customers.