Skip navigation

AMD Business

10 Posts authored by: scott.aylor Employee
scott.aylor

HPE and AMD: A Year Later

Posted by scott.aylor Employee Nov 27, 2018

What a year for the AMD and HPE teams! Only six months ago, I blogged about the AMD EPYC™ team heading to HPE Discover in Las Vegas and now we’re finishing up with HPE Discover in Madrid. It’s been a fantastic year for both companies with much more to come.

 

Before Las Vegas, we announced the HPE ProLiant DL325 Gen10 no compromise, single-socket server. This powerhouse platform from HPE is designed to tackle dense virtualization and software-defined storage workloads. With up to 32 cores, two terabytes of memory and 40 terabytes of NVMe storage in a 1U chassis, the DL325 is a fantastic machine for highly virtualized, on-premise workloads. The HPE ProLiant DL325 Gen10 delivers up to 27 percent lower cost per virtual machine (VM) than the leading dual-processor competitor for virtualization for those workloads[1].

 

Beyond the enterprise, we’ve seen some great wins for AMD and HPE in the high performance computing space, where AMD EPYC processors provide the necessary memory bandwidth and I/O lanes for memory intense workloads. At ISC in June 2018, HPE launched the new Apollo 35, a high density AMD EPYC™ compute solution that is ideal for memory bandwidth or capacity bound HPC workloads, such as computational fluid dynamics, weather simulation, and oil and gas exploration.

 

At SC18 in Dallas, HPE and AMD announced support for a new supercomputer from the High Performance Computing Center in Stuttgart, Germany. It will be the world’s fastest supercomputer for industrial production, powering computational engineering and research across science and industrial fields to advance applications in energy, climate, mobility, and health. Called Hawk, the supercomputer will be based on HPE’s next-generation high-performance computing platform running a next generation AMD EPYC™ processor code named “Rome.”

 

It's been a fantastic year for AMD and HPE. If you’re at HPE Discover Madrid, this is where you can find myself and the team on the show floor:

  • Stop by our booth, #230, to meet AMD experts and to see demos of AMD EPYC processors and HPE servers for virtualization, software-defined storage and more.

 

  • Tuesday, November 27 from 9:00 – 9:30 AM CET: HPE Live Interview with myself and Tom Lattin, VP, HPE ProLiant and Cloudline Systems.

 

  • Tuesday, November 27 from 10:00 – 10:30 AM CET at Theater 6: Olivier Suinat, CVP Sales, AMD Datacenter Solutions Group, will be on a panel discussing Cloud 28+ and AMD’s participation in it.

 

  • Wednesday, November 28 from 11:00 AM – 12:00 PM CET at N116, NCC Level 1: Tom Lattin, VP, HPE ProLiant and Cloudline Systems, will join me in a breakout session where we will talk about how HPE servers with AMD EPYC processors can redefine virtualization, software-defined storage and high-performance computing.

 

  • Wednesday, November 28 from 2:00 – 3:00 PM CET at N118, NCC Level 1: Isidro Gonzalez, senior GPU market development manager, AMD, will discuss AMD GPUs in an HPE GPU-enabled data center.

 

[1] Based on a comparison of the SPECvirt_sc2013 results of the ThinkSystem SR650 with 2 Intel Xeon Platinum 8164 processors versus the HPE ProLiant DL325 Gen10 with 1 AMD EPYC 7551P. SPEC and the benchmark name SPECvirt_2013 are registered trademarks of the Standard Performance Evaluation Corporation (SPEC). The stated results are published as of 06-05-18; see spec.org. Lenovo pricing from Lenovo site as of 05-14-18. HPE pricing is internal as of 06-05-18. Based on HPE testing, not independently verified by AMD.

The information contained herein is for informational purposes only and is subject to change without notice. Timelines, roadmaps, and/or product release dates shown in these slides are plans only and subject to change. “Rome is a code name for AMD architecture, and is not a product name. GD-122

 

Scott Aylor is the CVP & GM of  the AMD Datacenter Solutions Group. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.  GD-5

 

This week at the Open Compute EU Summit in Amsterdam I had the pleasure of rolling out along with ZT Systems and Microsoft, a powerful new addition to the AMD EPYC processor ecosystem. A leading provider to hyperscale datacenters, ZT Systems delivers outstanding performance and value for GPU computing and Virtual Desktop Infrastructure (VDI) with their new expansion system built on AMD Radeon Instinct GPUs. The support of Microsoft and its Project Olympus platform is the underlying foundation that ties the solution together.

 

Scott Aylor_ZT Systems.jpg

Enterprises are quickly moving some or all visualization workloads to the cloud.  Be it classic workstation workloads delivering design and manufacturing applications for CAD/CAE or just plain old Windows 10. Being able to deliver high-end applications and Windows desktops from servers in the cloud lets modern business bemore efficient and nimbler, with greater control.

 

The XPO200 3U PCIe Expansion System is designed to deliver outstanding flexibility and scale-out GPU computing, as well as highly cost effective VDI solutions, via AMD Radeon Instinct MI25 GPUs and EPYC CPUs.

 

AMD’s GPUs for the cloud have been purpose-built to optimally virtualize and deliver these workloads.  Using a built-in virtualization engine, based on the PCI spec SR-IOV (Single Root IO Virtualization), the GPUs deliver persistent quality of service making it easier to deploy and manage.

 

With up to 32 cores and 64 threads, 8 memory channels with up to 2 TB of memory per socket, and 128 PCIe lanes, AMD EPYC CPU is designed  to deliver the connectivity, memory capacity and VM density for GPU accelerated workloads.

 

To learn more about the XPO200 3U PCIe Expansion System as an attractive option for an array of modern datacenter applications please visit www.ztsystems.com/ocp.

 

The information contained in this blog represents the view of AMD or the third-party presenter as of the date presented. AMD and/or the third-party presenters have no obligation to update any forward-looking content in the above presentations. AMD is not responsible for the content of any third-party presentations and does not necessarily endorse the comments made therein. GD-84

 

Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied. GD-97

 

©2018 Advanced Micro Devices, Inc.  All rights reserved. AMD, the AMD Arrow logo, EPYC, Radeon, and combinations thereof are trademarks of Advanced Micro Devices, Inc.  Other product names used in this publication are for identification purposes only and may be trademarks of their respective companies.

As vast as the datacenter market is, it’s a relatively short list of companies working together in the day-to-day business. I don’t typically have the pleasure of engaging closely with a company that literally has hundreds of millions of customers like Dropbox. With over 500 million users and 300,000 Dropbox Business customers accessing its global collaboration platform, Dropbox is the latest big name in cloud to deploy the AMD EPYC™ processor in their custom-built infrastructure. 

 

“AMD EPYC is a compelling processor option for our compute technology, providing Dropbox with the technical specifications required to support the workloads that matter to teams and our individual users,” said Rami Aljamal, Head of Hardware Engineering and Supply Chain at Dropbox. “We are excited to deploy EPYC processors and look forward to working closely with AMD in the future.”

Dropbox.png

Dropbox will leverage AMD EPYC™ 7351P one-socket processor platforms to support future growth beyond its current capabilities and refresh its existing infrastructure for its most demanding compute workloads.

 

The AMD EPYC™ 7000 series delivers compelling options for the Dropbox offering, meeting performance demands throughout evaluation, qualification and deployment.  With 16 high-performance cores on the EPYC 7351P processor and leading-edge memory bandwidth, AMD continues to drive a strong balance of compute and connectivity while eliminating the need for a second socket. 

The AMD EPYC™ team is on the road again, and the next stop is HPE Discover. I’m headed to Las Vegas next week on the heels of our latest announcement, delivered by Dr. Lisa Su at Computex earlier this month, that AMD EPYC processors are at the core of the new HPE ProLiant DL325 Gen10 no compromise, single-socket server.

 

128832_EPYC_HPE_DL325Announcement_TW_1024x512_FNL.jpgIn case you missed it, this powerhouse new platform from HPE is designed to tackle dense virtualization and software-defined storage workloads. With up to 32 “Zen” processor cores from AMD EPYC, two terabytes of memory and 40 terabytes of NVMe storage in a 1U chassis, along with 40 PCIe lanes dedicated up to 10 NVMe drives, the HPE ProLiant DL325 Gen10 delivers up to 25 percent lower cost per virtual machine (VM) than the leading dual-processor competitor for virtualization[1]. Take a peek under the hood at our AMD Booth #146.

 

The HPE ProLiant DL325 Gen10 is one of a number of AMD-based HPE products we will be showing off next week. You can also check out the two-socket HPE ProLiant DL385 Gen10, and HPE MicroServer, as well as a live virtual desktop infrastructure (VDI) demo from the cloud.

 

In addition to the demos and action taking place at the AMD booth, here are a few other events within HPE Discover where you can find me and the team on the show floor:

 

  • Tuesday, June 19 from 8:30-9:00 AM CT: HPE Live interview with myself, Justin Hotard, Vice President and General Manager, Volume Global Business Unit, HPE.

 

  • Wednesday, June 20 from 9:00-10:00 AM PT: Breakout session with myself, Tom Lattin, Vice President, Server Options, HPE and Paul Brenner, Associate Director, Center for Research Computing, University of Notre Dame. Learn how AMD EPYC + HPE are building the foundation of your resilient enterprise, focusing on the HPE ProLiant DL385 and its security, agility and economic advantages for virtualization and software-defined storage environments. Register here to attend and receive location details.

 

  • Wednesday, June 20 from 11:00 AM-12:00 PM PT: Breakout session with Devin Fantasia, Senior Market Development Manager, Datacenter and Embedded Solutions Group, AMD, to dive deeper into the brand new HPE ProLiant DL325 Gen10 single-socket server that delivers 2P performance with 1P economics and efficiency[2]. Register here to attend and receive location details.

 

  • Wednesday, June 20 from 3:00-4:00 PM PT: Breakout session with Isidro Gonzalez, Senior GPU Market Development Manager, AMD. Hear about how AMD graphics processing units (GPUs) in HPE server platforms drive superior solutions for desktop virtualization, application virtualization and machine learning. Register here to attend and receive location details.

 

Be sure to keep up with us on Twitter @AMDServer for more on-the-ground updates. Hope to see you here!

 


[1] Based on a comparison of the SPECvirt_sc2013 results of the ThinkSystem SR650 with 2 Intel Xeon Platinum 8164 processors versus the HPE ProLiant DL325 Gen10 with 1 AMD EPYC 7551P. SPEC and the benchmark name SPECvirt_2013 are registered trademarks of the Standard Performance Evaluation Corporation (SPEC). The stated results are published as of 06-05-18; see spec.org. Lenovo pricing from Lenovo site as of 05-14-18. HPE pricing is internal as of 06-05-18. Based on HPE testing, not independently verified by AMD.

[2] Based on a comparison of SPEC CPU 2017 results of the Dell EMC PowerEdge R440 with 2 Intel Xeon Gold 5188 versus the HPE ProLiant DL325 Gen10 with 1 AMD EPYC 7551P. SPEC and the names SPEC CPU and SPECrate are registered trademarks of the Standard Performance Evaluation Corporation (SPEC). The stated results are published as of 06-05-18; see spec.org and http://h20195.www2.hpe.com/V2/GetDocument.aspx?docname=a00047709enw. Dell pricing from Dell site as of 05-16-18. HPE pricing is internal as of 06-05-18. Based on a comparison of the SPECvirt_sc2013 results of the ThinkSystem SR650 with 2 Intel Xeon Platinum 8164 processors versus the HPE ProLiant DL325 Gen10 with 1 AMD EPYC 7551P.  SPEC and the benchmark name SPECvirt_2013 are registered trademarks of the Standard Performance Evaluation Corporation (SPEC). The stated results are published as of 06-05-18; see spec.org. Lenovo pricing from Lenovo site as of 05-14-18. HPE pricing is internal as of 06-05-18. Based on HPE testing, not independently verified by AMD.

Today, I am proud to announce yet another first for EPYC: The new, density-optimized Cisco UCS C4200 Series Rack Server Chassis and the Cisco UCS C125 M5 Rack Server Node will be powered by AMD EPYC™ 7000 series processors. By integrating AMD EPYC processors, Cisco joins a growing list of server providers taking advantage of the high-performance EPYC processor’s strong balance of core density, memory, I/O bandwidth and unprecedented security features to deliver revolutionary technology for customers.

 

KT15053.png

The C4200 Rack Server Chassis hosts up to four rack server nodes in two rack units (2RU) with shared N+1 redundant power and cooling, and up to 24 small form factor drives.  The UCS C125 M5 Rack Server Node supports up to 2 AMD EPYC™ processors with up to 32 cores per processor, up to 2 TB of memory, two PCIe® 3.0 slots, and an optional 4th generation Cisco UCS VIC for complete programmability, making hosting different workloads simple and easy.  The result is Cisco’s highest density solution designed for service providers building cloud platforms, manufacturers simulating new designs, retailers analyzing consumer trends, compute-intensive web and gaming back-end processing and data scientists analyzing financial markets.   In fact, with 128% more cores, 50% more servers, and 20% more storage per rack1, the Cisco solution is designed for all clustered workloads where high core density is essential.

 

On the management side, this solution is fully supported by Cisco Intersight™ management as a service. This cloud-based management approach lets administrators configure and manage all Cisco blade, rack, storage, and multi-node servers through a single interface, regardless of where the servers are installed. This role and policy-based management service enables the creation of Cisco UCS service profiles and templates that have global scope across your organization worldwide.  Critical to service providers, secure multi-tenancy is enabled by integrating a cluster through Cisco Application Centric Networking (Cisco ACI™), helping to securely partition multiple workloads with network profiles that isolate tenants and applications.

 

“The addition of AMD to the Cisco UCS server portfolio marks a first for us as partners,” said Kaustubh Das, vice president, product management - Computing Systems, Cisco’s Data Center Business Group. “Leveraging the innovation of AMD EPYC, Cisco is bringing forth transformative technology that will enable our customers to accelerate compute-intensive workloads with a high-density server that can be managed from the cloud."

 

The Cisco UCS C4200 Series Rack Server Chassis and C125 Server Nodes with AMD EPYC processors are expected to be available in the second half of 2018 with tested and validated solutions with major ISVs.

 

If you would like to learn more about these great new products, please visit the links below:

 

1https://www.cisco.com/c/en/us/products/servers-unified-computing/ucs-c4200-series-rack-server-chassis/index.html

Recently, during a larger briefing on AMD in the Enterprise, I gave an AMD EPYC™ processor update to press and analysts on our progress since we launched last June. It was satisfying to pause for a few minutes, take stock, and reflect on the amazing momentum we’ve created in less than a year. A special guest joined me half way through with some exciting news that shows we’re just getting started on the journey in bringing value to customers through the innovations on EPYC™ processors. Below are some of the highlights from the session.

 

I started by giving a quick update on our ramp to date. Bottom line, a month away from our first birthday, we’ve been aggressive and purposeful in ramping the EPYC™ processor business, helping the industry rethink the server and the datacenter.

 

  • We’ve built a revolutionary product line eagerly adopted by world class customers, ecosystem partners, and their customers. In fact, we now have 14 server systems partners and over 50 server platforms introduced and ramping.
  • We’ve focused on rethinking the technology and economics of the server to help our customers win in the era of the software designed data center. The results are in: for virtualized environments, EPYC™ based systems can deliver up to 28% lower total cost of ownership than similarly configured Intel Xeon servers.
  • Equally important, we’ve created superior performance, value, and advanced security features in emerging, rapidly growing use cases. Here’s some of the compelling examples I walked through. I’ve put some links at the bottom of this in case you want to get your ‘geek on’ and go deep:

 

See Endnotes

 

One of the amazing customer examples I walked through was, Hivelocity. A dedicated server and cloud hosting provider, they’ve already taken the leap in deploying what we’ve coined the “industry’s first no-compromise single server” approach.  It’s paid off, according to their COO Steve Eschweiler: “Our AMD EPYC™ processor-powered Tyan servers have truly blown away our price/performance expectations”. In fact, with a single AMD EPYC processor the Hivelocity client is experiencing read/write speeds six times faster than that of two Xeon processors.

 

Quite an accomplishment. We’re proud, and thankful… But we’re just getting started.

 

I was then joined by a special guest Lee Caswell, VP of Products, Storage and Availability at VMware. Lee joined me to cover the next high-growth use case we’re rethinking: Hyperconverged Infrastructure (HCI for short).  That same ‘no compromise’ single socket approach applied by Hivelocity has the potential of putting this hypergrowth market into warp speed.

 

For those that aren’t familiar with HCI, it is fundamental to the Software Defined Datacenter. HCI uses software-defined storage through the same server resources used to run Virtual Machines, eliminating legacy storage systems, converging all elements into a single, easily managed, pool. VMware vSAN™ runs on industry-standard x86 servers and components that help lower TCO by up to 50% versus traditional storage. Lee helped me walk through the market, the challenges, and the opportunities through the lens of the HCI industry leader, VMware vSAN. Some takeaways from Lee’s presentation:

 

  • HCI is a $5B market growing approximately 30% year over year. And, VMware vSAN is the undisputed software leader according to IDC’s latest report on April 3 of this year.
  • Customers are adopting HCI not only for the prospect of massively reducing total cost of ownership, but also to accelerate the transformation to software defined data centers (SDDC) and move towards more flexible cloud consumption models.
  • To meet customer requirements for HCI, configuration and testing are rigorous and deep. This goes well beyond just a standard Hardware Compatibility List (HCL). Many customers prefer to consume HCI as an appliance like Dell EMC’s VxRail as a result.
  • But, many customers prefer to build their own HCI systems. This is where VMware’s proven vSAN ReadyNode program has tapped into an unmet need. These servers, precisely configured and rigorously tested, help ensure that customers can be confident in their ability to quickly deploy, scale, and operate HCI clusters.  In fact, Lee mentioned that vSAN ReadyNodes actually make up the majority of vSAN implementations.
  • Finally, most vSAN ReadyNodes to date have had to be beefy two socket platforms that are ripe for optimization and new thinking for storage centric environments. That’s where VMware, Dell, and AMD EPYC are really excited about the world’s first EPYC™ processor-based VMware certified vSAN ReadyNode.

 

The AMD EPYC™ processor-powered Dell PowerEdge 7415 recently launched and received its vSAN ReadyNode certification along with its big and little brother the PowerEdge 7425 and 6415. As a “no compromise” single socket server, it was designed with the prospect of delivering leading core density and I/O capability for storage heavy HCI use cases. Well, once again, the results are in. And they’re EPYC™. Lee and I revealed recent third-party test results showing we’ve more than delivered on that potential. In fact, versus currently shipping Xeon based systems, Dell shows the EPYC processor-powered PowerEdge 7415 delivering up to 20% lower TCO and slashing licensing costs by up to half.  This is EPYC.

 

See Endnotes

Thank you to Lee, his team at VMware, and the Dell EMC team for the partnership.

 

To close, I’d like to reiterate that we’re only getting started, expect to see more partners, more platforms, more performance, more value, more security in the coming months as we ramp to our first birthday and beyond.  This is going to be EPYC™.

 

 

28% Lower TCO based on 3-year Virtual Machine Cost: 1. Data received from HPE TCO Calculator. Configure your own analysis at https://www.hpe.com/us/en/solutions/tco-calculators.html. 2. Based on an estimate of 1.5 virtual machines per core, the greater number of core in AMD EPYC processors exceeds that available in Intel Xeon Scalable processors 3. Based on comparison of top published SPECrate2017_fp_base scores for HPE ProLiant DL380 Gen10 and DL385 Gen10. AMD EPYC results available at www.spec.org as of 4/29/2018. 4. Source of pricing Configure your own TCO analysis https://roianalyst.alinean.com/ent_02/AutoLogin.do?d=898755097515045746. HPE ProLiant DL380 Gen10 config: 1 x 6130 (16 Total Cores), 64GB Memory, 8 SFF Chassis, 2x800W PS, P408i-a;  HPE ProLiant DL385 Gen10 config: 1 x 7451 (24 Total Cores). 64GB Memory, 8 SFF Chassis, 2x800W PS, P408i-a

 

25% Fewer Servers for In Memory Analytics: Tests commissioned by AMD and run by Principled Technologies, Inc. running HiBench Spark, processing a 220 GB k-means dataset. AMD cluster of: 3 x Grandstands (chassis est. $600 ea.), each with (2) AMD EPYC 7601 SOC (AMD 1ku pricing $4200 ea), RAM: 32 x 16GB sticks ($169 ea. Samsung M393A2K40BB1-CRC per newegg.com) of DDR4 (512GB of RAM per server, 1536GB per cluster), Disk: 24 x Samsung MZ-7LM120E 120GB SSDs ($113 ea. per CDW), Network: 1 x Mellanox ConnectX-4 Lx ($332.77 per newegg.com); (for total cluster cost of approx. $52,449), completed the test in 213.6 seconds; versus Intel cluster of:  4 x Lenovo servers (chassis est. $600 ea.), each with (2) Intel Xeon E5-2699 v4 ($4115 ea. per ark.intel.com), C612 chipset ($54 per ark.intel. com), RAM: 24 x 16GB sticks ($169 ea. Samsung M393A2K40BB1-CRC per newegg.com) of DDR4 (384GB of RAM per server, 1536GB per cluster), Disk: 24 x Samsung MZ-7LM120E 120GB SSDs ($113 ea. per CDW), Network: 1 x Mellanox ConnectX-4 Lx ($332.77 per newegg.com), HBA: 1 x LSI LOGIC 9305-16i HBA card ($399.99 per newegg.com); (for total cluster cost of approx. $65,628), completed in 243.1 seconds. Each cluster was connected via a Mellanox SN2700 model switch, and ran Red Hat Enterprise Linux 7.3, Hortonworks Data Platform (HDP) 2.6, and OpenJDK 1.8.  NAP-23 Testing by Principled Technologies and not verified by AMD. This compares acquisition cost for a 4-node cluster of Intel Xeon E5 2699 v4 processor-based servers compared to a 3-node cluster of AMD EPYC 7601 processor-powered servers.  NAP-23

 

50% More Throughput for NoSQL Database: AMD EPYC 7601 based server performance comparison to Intel Xeon 2699 v4 based server in Principled Technologies white paper at http://www.principledtechnologies.com/AMD/EPYC_ Cassandra_competitive_1217.pdf

 

77% More Performance for Weather Research Forecasting: AMD internal testing on WRF v3.8.1 (GCC 5.4.0 compiler with Basic Nesting (WRF option) and MPI 3.2 (dmpar compile flag); CONUS 12KM dataset; WRFIO_NCD_LARGE_FILE_SUPPORT=1). Tests conducted on AMD Grandstand reference system with 2 x EYPC 7601 SoCs, 16 x 32GB DDR4-2400MHz, 3 x 350GB SATA SSDs, Ubuntu 16.04.2 completed the test in an average of 65 seconds; versus HPE DL380 server with 2 x E5-2699A v4 processors, 16 x 16GB DDR4-2400MHz, 3 x 1TB NVMe and 3 x 1TB SATA SSDs, Ubuntu 16.04.4 completed the test in an average of 115 seconds. NAP-92

 

133% Better Performance/Dollar for Supercomputing: Based on SPECfp®_rate2006 scores published on www.spec.org as of October 25, 2017.  2 x EPYC 7601 CPU ($4,200 per processor at AMD 1ku pricing) in Sugon A620-G30, Ubuntu 17.04, x86 Open64 v4.5.2.1 Compiler Suite, 512 GB PC4-2666V-R memory, running at 2400  1 x 1TB SATA 7200RPM has a peak score of 1850 (base score 1670); versus 2P Xeon Platinum 8180 ($10,009 per processor per ark.intel.com)-based Huawei 2288H V5 system with SUSE Linux Enterprise Server 12 SP2, ICC 17.0.0.098, 384GB PC4-2666V-R memory, 1x1200GB SAS 10000RPM score of 1890 (base score 1850).  SPEC and SPECfp are registered trademarks of the Standard Performance Evaluation Corporation.  See www.spec.org for more information. NAP-47

26% Faster Simulations for Computational Fluid Dynamics:  Based on Dell internal testing using the ANSYs Fluent benchmark test in November 2017, comparing to a similarly configured Dell PowerEdge R7425 with a traditional processor. Actual performance will vary based on configuration, usage and manufacturing variability. World Record SPECfp(R)_rate2006: Result available at https://www.spec.org/cpu2006/results/res2017q4/cpu2006-20171114-50603.html. 2 x EPYC 7601 CPU in HPE ProLiant DL385 Gen10, SUSE Linux Enterprise Server 12 SP3, x86 Open64 v4.5.2.1 Compiler Suite, 1 TB (16 x 64GB 4Rx4 PC4-2666V-L) memory, 1 x 300 GB 15k RPM SAS.  SPEC and SPECfp are registered trademarks of the Standard Performance Evaluation Corporation.  See www.spec.org for more information.

World Record SPECrate2017_fp_peak as of May 21, 2018: Result available at https://www.spec.org/cpu2017/results/res2018q2/cpu2017-20180319-04087.html. Based on SPEC CPU 2017 scores published on April 19, 2018. 2 x EPYC 7601 CPU in Supermicro A+ Server 4023S-TRT, SUSE Linux Enterprise Server 12 SP3 (x86_64) kernel 4.4.114-94.11-default, AOCC v1.0.0 compiler, 1 TB (16 x 64GB 4Rx4 PC4-2666V-L) memory, 1 x 500 GB SATA III 7200RPM has a peak score of 279 (base score 267). NAP-96

Up to 50% lower TCO VMware vSAN 6.7 Datasheet at https://www.VMware.com/content/dam/digitalmarketing/VMware/en/pdf/products/vsan/VMware-vsan-datasheet.pdf

Up to 20% Lower TCO and up to 50% Lower Licensing Costs: Demartek test report: Dell EMC PowerEdge R7415 AMD EPYC VMware vSAN Mixed Workloads Performance – April 2018 www.demartek.com/Demartek_Dell_EMC_PowerEdge_R7415_vSAN_Mixed_Workloads_Evaluation_2018-04.html

I spent this week at Dell Technologies World, and they sure know how to put on a show. The energy was palpable, with jam-packed keynotes, dynamic guru speakers and interactive breakout sessions. Dell truly is “making it real” and we are pleased to be a key technology partner along this journey with our EPYC™ single and dual socket platforms.

 

DTW Booth 2.jpgWith more than 14,000 attendees at this year’s event, our AMD booth experienced heavy traffic, with all eyes on our “AMD is EPYC” campaign highlighting Dell EMC PowerEdge products. We showed off a series of demos that underscored why EPYC is the ideal foundation for a variety of workloads including HPC (25% greater performance for workloads like Computational Fluid Dynamics[i]), database management, VM migration (simplified integration with VMware vSphere), data analytics (faster Apache Spark™ with 25% fewer servers[ii]) and more.

 

On Tuesday, I hosted a breakout session that focused on the compute, memory, I/O and security advantages of our EPYC CPUs + Dell EMC PowerEdge servers for software-defined, virtualized and high-performance computing (HPC) workloads in today’s datacenter.

 

The session covered everything from the current datacenter landscape and IT challenges with incrementalism, to how we re-entered the market with a fresh approach and unparalleled features, to the specific advantages that EPYC processors provides for target workloads, to the head-to-head comparison of both the single socket Dell EMC PowerEdge R6415 and Dell EMC PowerEdge R7415 servers against relative Intel-based platforms. The results did the talking, with the EPYC 7351 processor in the Dell EMC PowerEdge R6415 delivering 33% more memory per processor[iii] and 33% more cores[iv] than the competition, and the EPYC 7601 powering the Dell EMC PowerEdge R7415 offering 33% more memory per processor[v] and 14% more cores[vi] than competitive offerings, and according to Dell offers up to 20% better TCO per four-node cluster for vSAN deployments at the edge[vii]. Together with Dell EMC PowerEdge, we are redefining the rules of the modern datacenter, improving performance and adding competitive value for our customers.

 

AMD really is striking at the heart of the market with EPYC processors, and the past few days have reinforced our commitment to the EPYC portfolios of today as well as primed audiences for our plans for future generations of groundbreaking products. We are helping OEMs, cloud companies and ecosystem partners to “break the habit” of selecting the status quo and recognize the business value that we are driving with EPYC.

 

You can see more highlights and pictures from Dell Technologies World on our AMD Twitter and LinkedIn channels.


 


[i] “High Performance Computing” – the dual socket Dell EMC Poweredge R7425 delivers up to 24% improved performance vs. the HPE DL385 for containers, hypervisors, virtual machines, and cloud computing and up to 25% absolute performance improvement for HPC workloads like computational fluid dynamics [CFD].  With up to 64 cores, it offers high bandwidth with dense GPU/FPGA capability.  On standard benchmarks, the server with superior memory bandwidth and core density provided excellent results across a wide range of HPC workloads.” Data provided by ell Feb. 2018, not verified by AMD. Learn more at https://www.emc.com/about/news/press/2018/201802016-01.htm

[ii] Tests commissioned by AMD and run by Principled Technologies, Inc. running HiBench Spark, processing a 220 GB k-means dataset. AMD cluster of: 3 x "Grandstand" reference systems (chassis est. $600 ea.), each with (2) AMD EPYC 7601 SOC (AMD 1ku pricing $4200 ea), RAM: 32 x 16GB sticks ($169 ea. Samsung M393A2K40BB1-CRC per newegg.com) of DDR4 (512GB of RAM per server, 1536GB per cluster), Disk: 24 x Samsung MZ-7LM120E 120GB SSDs ($113 ea. per CDW), Network: 1 x Mellanox ConnectX-4 Lx ($332.77 per newegg.com) - for total cluster cost of approx. $52,449 - completed the test in 213.6 seconds; versus Intel cluster of:  4 x Lenovo servers (chassis est. $600 ea.), each with (2) Intel Xeon E5-2699 v4 ($4115 ea. per ark.intel.com), C612 chipset ($54 per ark.intel.com), RAM: 24 x 16GB sticks ($169 ea. Samsung M393A2K40BB1-CRC per newegg.com) of DDR4 (384GB of RAM per server, 1536GB per cluster), Disk: 24 x Samsung MZ-7LM120E 120GB SSDs ($113 ea. per CDW), Network: 1 x Mellanox ConnectX-4 Lx ($332.77 per newegg.com), HBA: 1 x LSI LOGIC 9305-16i HBA card ($399.99 per newegg.com)- for total cluster cost of approx. $65,628 - completed in 243.1 seconds. Each cluster was connected via a  Mellanox SN2700 model switch, and ran Red Hat Enterprise Linux 7.3, Hortonworks Data Platform (HDP) 2.6, and OpenJDK 1.8.  NAP-23

[iii] AMD EPYC™ 7351 processor supports up to 8 channels of DDR4-2400, versus the Xeon Gold 5118 processor at 6 channels of DDR4-2400. NAP-42

[iv] AMD EPYC™ 7351 processor includes up to 16 CPU cores versus the Intel® Xeon® Gold 5118 processor with 12 CPU cores.

[v] AMD EPYC™ 7601 processor supports up to 8 channels of DDR4-2667, versus the Xeon Platinum 8180 processor at 6 channels of DDR4-2667. NAP-42

[vi] AMD EPYC 7601 processor includes up to 32 CPU cores versus the Xeon Platinum 8180 processor with 28 CPU cores. NAP-43

[vii] From Dell EMC Press release, “Dell EMC Expands Server Capabilities for Software-defined, Edge and High-Performance Computing”. The 2U single-socket Dell EMC PowerEdge R7415 is the first server platform with vSAN Ready Nodes and offers up to 20% better TCO per four-node cluster for vSAN deployments at the edge.  Based on Dell internal analysis in January 2017, comparing the projected list price cost of a similarly configured Dell PowerEdge R7415 single-socket server versus the actual list cost of a traditional dual-socket server. Actual costs will vary.

On June 20, 2017, AMD disrupted the server space in a big way with the announcement of our EPYC™ 7000 series processors, including support from a global ecosystem of customers and partners. Fast forward to the new year and AMD is celebrating the rejuvenation of its embedded solutions by welcoming a new processor to the family.

 

EPYCBlog.jpgThe performance of the “Zen” architecture and datacenter insight from EPYC are infused in the make-up of the company’s next-generation embedded CPU – the AMD EPYC Embedded 3000 processor family. These two distinct and important qualities, architecture and insight, position AMD to help partners and customers take a sophisticated approach to navigating and exploring the shift in computing from the server to the edge.

 

As a result, workloads no longer need to be solely dependent on the core of a network to process and complete intensive tasks. The AMD EPYC Embedded 3000 processor family meets the stringent requirements for the majority of embedded processors, including power management and dependability, while delivering on major advancements in I/O integration, flexibility and security. This advanced processor is ready to support network function virtualization (NFV), software-defined networking (SDN), industrial system applications and more. Additionally, the AMD EPYC Embedded 3000 processor family delivers up to 2.7X more performance-per-dollar[i] and up to 2X more connectivity[ii] than competitive solutions on the market today.

 

At a Glance: EPYC Embedded 3000 Processor Family

 

  • The AMD EPYC Embedded 3000 processor family is based on the ‘Zen’ architecture and a 14nm FinTET process
  • A wide range of core counts, including options for 16, 12, 8 and 4 cores per socket
  • Up to 64 PCIe Gen 3 lanes
  • High-performance single and multithreaded processing
  • Up to 8 channels of 10GbE Ethernet
  • Up to 32MB shared L3 cache with options for 4 independent memory channels
  • TDPs ranging from 30W to 50W (for 1 die and up to 8 cores) and 60W to 100W (for 2 dies and up to 16 cores)
  • Unparalleled enterprise-grade reliability, availability and serviceability (RAS) features
  • Product availability for up to 10 years, offering customers a long lifecycle support roadmap

 

The Ecosystem Perspective: Networking and Communications with Seagate

 

Well-positioned to tackle storage applications in dynamic new ways, Seagate, a world leader in storage solutions, is evaluating the AMD EPYC Embedded 3000 to support the company’s intelligent storage array technologies. With a growing abundance of storage information populated daily, Seagate customers are frequently searching for ways to break down dependability and performance challenges.

 

Seagate is particularly interested in next-generation performance and enterprise-class RAS features to push for an up to 40 percent performance improvement over prior generations of product architectures. The performance gains are no small feat, and Seagate is working to help ensure that customers get the performance they need at an improved total cost of ownership.

 

In addition, AMD continues to work with a variety of customers and partners to deliver processing excellency and impact to workloads across other key industries including networking and industrial applications.

 

Concluding the First Zen-to-Zen Journey

The AMD EPYC Embedded 3000 product family is a nice capstone on the first generation of ‘Zen’-based products, including Ryzen™, EPYC, and now Ryzen™ Embedded and EPYC Embedded. However, this processor is also unique in the ‘Zen’-to-‘Zen’ experience delivered from a family of products that span from the server and the edge. As industries move toward enabling machine learning, AI and internet of things at the edge, the AMD EPYC Embedded 3000 is ready to shape advanced embedded capabilities today and into the future.

 

Cautionary Statement

This blog contains forward-looking statements concerning Advanced Micro Devices, Inc. (AMD) including the features, functionality, availability, timing, deployment, and expected benefits of the Ryzen™ Embedded V1000 processor and the products being developed by customers based on Ryzen™ Embedded V1000 processor as well as the expected support from major ecosystem partners, which are made pursuant to the Safe Harbor provisions of the Private Securities Litigation Reform Act of 1995. Forward-looking statements are commonly identified by words such as "would," "intends," "believes," "expects," "may," "will," "should," "seeks," "intends," "plans," "pro forma," "estimates," "anticipates," or the negative of these words and phrases, other variations of these words and phrases or comparable terminology. Investors are cautioned that the forward-looking statements in this document are based on current beliefs, assumptions and expectations, speak only as of the date of this document and involve risks and uncertainties that could cause actual results to differ materially from current expectations. Such statements are subject to certain known and unknown risks and uncertainties, many of which are difficult to predict and generally beyond AMD's control, that could cause actual results and other future events to differ materially from those expressed in, or implied or projected by, the forward-looking information and statements. Investors are urged to review in detail the risks and uncertainties in AMD's Securities and Exchange Commission filings, including but not limited to AMD's Annual Report on Form 10-K for the year ended December 30, 2017.



[i] Estimates based on SPECint®_rate_base2017 using the GCC-02 v6.1 compiler. AMD-based system scored 24.2 in tests conducted in AMD labs as of 12 January, 2018, configured with 1 x EPYC 3251 SOC ($315 per processor at AMD 1ku pricing), 32GB memory (2 x 16GB 2Rx4 PC4-2666 running at 2666), 1x 250 GB SSD, AMD Wallaby Rev C, RHEL 7.4. Intel Xeon D 1540 scored 16.1. based on tests conducted in AMD labs as of 12 January, 2018 using Supermicro Server System X10SDV-8C-TN4F, configured with 1 x Xeon D 1540 ($581 each processor per ark.intel/com), 32GB memory (2 x 16GB 2Rx4 PC4-2666 running at 2133),  1x 250 GB SSD, RHEL 7.4.EMB-152.

[ii] AMD EPYC™ Embedded 3451 supports up to 64 PCI Express high spend I/O lanes, 8 10 GbE, 16 SATA, and 4 memory channels versus Xeon D 1587 supports 32 PCIe lanes, 4 10GbE, 6 SATA, 2 memory channels. EMB-153.

It has been quite the month for datacenter news across the industry. With a lot of information flying around, we want to take a moment to engage directly and answer questions regarding our EPYC™ 7000 series of high-performance server processors, launched in June and designed to once again deliver AMD innovation to the datacenter.

 

We invite you to join Forrest Norrod, SVP and GM of Enterprise, Embedded & Semi-Custom Products (EESC), Kevin Lepak, AMD Fellow, and me, Scott Aylor, CVP and GM, Enterprise Solutions on Tuesday, July 18 from 12:00-1:00 pm CT for a Facebook Live Q&A where we will be addressing questions from the community about our “Zen” architecture, EPYC processor performance and workload metrics, and our robust ecosystem of OEMs, ODMs, cloud providers and hardware partners that stand behind EPYC. We’ll also be available to answer other technical inquiries on what this new processor means for the datacenter.

 

As a quick reminder, you can keep up with all things EPYC here.

 

For more background, below are links to a comprehensive list of EPYC processor information resources, including white papers, AMD Datacenter Tech Day presentations, videos, Launch Day executive presentations and more. You can also access this information in one place here.

 

We look forward to continuing the EPYC conversation!

 

--------------------------------------------------------------

Performance

  • “AMD EPYC SoC Breaks Records with SPEC CPU Benchmarks” Performance Brief
  • “AMD EPYC SoC Delivers Exceptional Results on the STREAM Benchmark on 2P Servers” Performance Brief
  • “EPYC: Designed for Effective Performance” Whitepaper
  • “Power / Performance Determinism” Whitepaper
  • “SpecCPU 2017 & Changing Performance” Whitepaper

Architecture

Disruptive Technologies

  • “AMD EPYC Empowers Single Socket Servers” Whitepaper
  • “Trusting in the CPU: Getting to the Roots of Security” Whitepaper
  • SEV and VM Boot Security Video
  • “AMD EPYC Empowers Server GPU Deep Learning” Whitepaper

Partner/Customer Ecosystem

“AMD EPYC SoC Breaks Records with SPEC CPU Benchmarks”

AMD’s Embedded technology expands the possible across various market segments, from education to casino gaming to office collaboration to healthcare. The latest example is Sony’s Future Lab concept “T”, a projector prototype that creates interactive space on a table or the surfaces of real objects.

 

Future Lab Concept T.jpgBy applying advanced machine learning techniques and a highly accurate depth-sensing system, “T” can recognize objects and gestures, transforming the way users interact in a variety of applications. As part of Sony’s Future Lab Program, announced at IFA this year, “T” will be opened up for development with select partners to evaluate new use cases in different situations over the coming months, leveraging feedback to refine and evolve the project.

 

“T” is powered by an AMD Embedded processor, which provides leading graphics and image processing in a power-efficient package. In addition to enabling high-performance applications, AMD’s power-efficient technology is particularly beneficial for projectors, helping keep heat and energy costs low.

 

The potential use cases for “T” span showrooms, shop fronts, office spaces and much more. For example, “T” can be used for educational simulation, where users move around different 3D objects on the tabletop to bring ideas to life in a simple, visual and tangible way. What has previously been sketched out on whiteboards during classroom hours moves beyond 2D, dramatically increasing interaction, and inspiring new ways to communicate and collaborate.

 

AMD is proud to support Sony’s Future Lab concept “T” with its embedded technology and we look forward to the next wave of innovation!

 

Scott Aylor is Corporate Vice President and General Manager for AMD Enterprise Solutions. His postings are his own opinions and may not represent AMD’s positions, strategies or opinions. Links to third party sites are provided for convenience and unless explicitly stated, AMD is not responsible for the contents of such linked sites and no endorsement is implied.