Skip navigation

We are excited to be at SC’19 with our friends and family of ecosystem partners. I’d like to share my thoughts on how AMD has unleashed the EPYC revolution for HPC. AMD is all about innovation and our mission is to deliver products that help to solve the world’s toughest challenges – in life sciences, earth science, energy, manufacturing, fundamental research, oil and gas, machine intelligence and many more. We celebrated our 50th anniversary milestone this year with what analysts called the ‘7nm storm’. The 7nm EPYC, Radeon and Ryzen processors bring new possibilities to the new era of computing with ground-breaking performance and outstanding power efficiency driving lower TCO.


Creating an inflection point with trailblazing performance and unprecedented scalability for today’s HPC workloads, AMD EPYC processors mark the next milestone in “exascale computing” characterized by compute power in exaFLOPS, or a quintillion floating-point calculations per second. AMD is uniquely positioned to lead the exascale era with CPU and GPU technologies. We are collaborating with the US Dept of Energy, Cray and Oak Ridge National Laboratory to build the world’s fastest supercomputer named Frontier, expected to hit 1.5 exaflops. This will be five times faster than today’s top supercomputers. Powered by AMD CPUs and GPUs, Frontier will help model the entire lifespan of a nuclear reactor, uncover disease genetics, and build on recent developments in science and technology to further integrate artificial intelligence with data analytics and modeling and simulation.


HPC touches every aspects of lives. HPC in the enterprise segment also is being accelerated as many industries are looking for faster and safer solutions for real world problems, challenging the status quo to find breakthrough innovations in fields such as weather modeling and simulation, materials and manufacturing industries, oil and gas, healthcare and medicine, to name a few. HPC requires high performance CPUs.


HPC is all about high performance CPUs. AMD EPYC offers a range of processor options for HPC. Let me highlight two specific CPUs from our broad portfolio of processors. EPYC 7542, with 32 cores (2.9GHz base, up to 3.4GHz boost, 225W TDP) and 128MB of L3 cache, has been a popular option in the middle of the market, while EPYC 7742, with 64 cores (2.25GHz base, up to 3.4GHz boost, 225W) and 256MB of L3 cache, has been a popular choice at the high end. New addition to our innovative portfolio is the EPYC 7H12 which packs 64 cores (2.6GHz base, up to 3.3GHz boost, 280W TDP) specifically built for extreme performance. Here are some examples of how AMD EPYC steps up the game, yet again. Our ecosystem partners have announced highly optimized server platform for HPC to address the performance and scalability needs of emerging demands.


Faster Weather Forecasting

We are reminded of the importance of weather forecasting every day. AMD EPYC empowers solutions to more efficiently predict weather, including weather-related natural disasters, which helps reduce the enormity of losses caused by these disasters. 


The Weather Research and Forecasting (WRF) Model is a popular application for predicting weather. It is used for both atmospheric research and operational weather forecasting applications. It’s data assimilation system and parallel compute capability allows WRF to server a wide range of meteorological applications.


AMD EPYC demonstrates exceptional performance and scalability running WRF and AMD EPYC 7742 has been a popular choice for it. With 128 cores and 256 threads in dual CPU configurations EPYC 7742 powered servers have demonstrated approximately twice the performance of our previous generation of EPYC processors. Since WRF is open source, there are no software license costs to consider in choosing the number of cores that you run.

See additional 2nd Gen AMD EPYC performance test reports running WRF use cases here.


Building Faster Physical Models through Computational Fluid Dynamics

Computational Fluid Dynamics (CFD) is another critical workload for solving today’s engineering challenges. We have tested several CFD codes and demonstrated industry leading performance on AMD EPYC 7002 series of processors. I want to highlight ANSYS CFX, a popular application which has a long history and is best known for its ability to simulate turbomachinery accurately and quickly.  Let us look at a performance of ANSYS CFX running on two mid-range SKUs – Intel Xeon Gold 6248 processor with 20 cores, 2.5GHz base frequency and 27.5MB cache, and, AMD 2nd Gen EPYC 7542 with 32 cores, 2.5GHX base frequency and 128MB of cache. 

On five standard ANSYS CFX benchmark models, the 2nd Gen AMD EPYC 7542 significantly outperforms the Xeon Gold 6248. Efficiently running this many cores per CPU with stellar results allows for much denser solutions.  More density with better performance allows reductions in total systems required resulting in, lower power, and a smaller footprint in the data center. 


Automotive Safety is Top of Mind

Driving a safe car is one of the highest priorities for consumers. Designing a safe car quickly is one of the highest priorities for automotive manufacturers. Designing better and safer products requires the engineers to predict the consequence of any design changes on the real-world performance of their product. 2nd Gen AMD EPYC allows car makers to analyze the safety of their designs faster, leading to safer cars and faster time to market. 


Altair Radioss is a leading structural analysis solver and has established itself as a leader and an industry standard for automotive crash, drop & impact analysis, terminal ballistic, blast and explosion effects and high velocity impacts.

Altair Radioss was used to compare the performance of the highest core-count 2nd Gen EPYC processor (AMD EPYC 7742) vs. the highest core-count industry-standard pin-based (LGA) competitive processor (Intel Xeon Platinum 8280).  We ran 2 standard benchmarks on both systems.  The results are summarized below.

Comparing the top of the product stack of 2nd Gen EPYC processors and Intel Xeon Platinum processors, once again demonstrates the dominant performance of the 2nd Gen EPYC processors.  The 7742 is 38% faster on average than the Intel Platinum 8280 across these two benchmark models.


See how AMD EPYC supports real world simulation for safety from the performance test results on Radioss.

2nd Generation EPYC processors are truly changing the game in HPC, delivering exceptional performance on real-world workloads.  Talk to your AMD sales team, your software partner, or your server partner to find out which AMD EPYC processor best fits your workload’s demands.  Innovation is in our DNA.  We are just getting started on the EPYC journey to revolutionize HPC!


We are grateful to our technology partners who have collaborated with our engineers in creating a wide range of datacenter application use cases:  Altair, Ansys, Atos, Broadcom, Cadence, Cray, Dassault Systems, Dell EMC, Docker, ESI Group, Gigabyte, HPE, LSTC, Mellanox, Mentor Graphics, Microsoft, Micron, Mentor Graphics, Microsoft, Oracle, Red Hat, Samsung, ScaleMP, Siemens PLM, Supermicro, SUSE, Synopsys, WekaIO, Xilinx and others.


Raghu Nambiar is the CVP & CTO of Datacenter Ecosystems & Application Engineering for AMD. 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. 

By Pawel Jaruga, 3D Artist, Poland



Creative technology for the ages


Over the course of my decade-long career as a 3D artist and digital sculptor, there has never been greater opportunity for creative freedom as there is today. With access to powerful software and hardware, professionals in creative industries today are able to put all of their efforts into bringing their vision to life – without being held back by technology.



At the start of my career, I found that my creative vision was often limited to creating and delivering what was easy, achievable and realistic. A decade on, and with several investments in technology along the way, my creative process – from modelling and texturing to shading and rendering – has allowed me to create increasingly detailed models faster and more seamlessly.


In the past, when processors (CPU) and graphic cards (GPU) didn’t have the high-performance capabilities they do today, I found that I was constrained by the number of triangles and faces I could achieve and the textures I could create in a character model.


This lack of processing power made things challenging when I was working from home using a single workstation as everything took so long, especially when creating animations. Even the difference in hardware now compared to three years ago has given my workflow a huge boost.


In my everyday workflow, I now use the 3D sculpting software Zbrush and several GPU renderers including AMD Radeon ProRender, while my hardware includes AMD Radeon Pro WX 9100 Graphics (provided to me by AMD) and an 8-Core 4.0 GHz processor with 64GB RAM. With previous hardware a few years ago, I was creating models where 20-50 million triangles models were considered a lot for a freelancer. But today, my workstation can handle upwards of 200 million triangles. So for high poly models going up to 70-120 million of triangles is nothing special anymore.


Meanwhile, for texturing, I use Substance Painter and 3D Coat and my hardware includes at least 8GB of video memory. This is really the minimum you can have for texturing a standard game character with a 4096-texture set, which means a resolution of 4096 x 4096 pixels. This is a huge jump from the start of my career, back when the texture limit was 1024 x 1024 per character.



Technology advancements such as these have not only allowed us creatives to develop more realistic and vivid characters, but have also enabled real-time rendering, which means I can see any changes as I make them, resulting in a much faster workflow. I also have the ability to produce photorealistic images that help create a highly detailed and believable 3D world.


You can also easily light scenes and models with single HDRI images and include photogrammetry scans in your modelling workflow – real world items and human 3-dimensional scans based on photo sets taken around objects, which can also include high resolution textures.


Overall, this helps make the creative process much more fun, as you’re not having to wrestle with your hardware. There really has never been a better time to be a video game designer.


Best ever performance, best ever value


The affordability of software and hardware tools has continued to improve, to support this technology journey toward creative freedom, and 3D artists and developers can now focus on using superior software and their skills to their full potential.



However, how you balance performance and value depends on your needs. My requirements change depending on whether I’m earning money from rendering work, or whether I’m modelling and texturing high-end video game characters where I need to be able to work in real-time with multiple 4K textures sets. Or whether I’m creating simple models for mobile games or lower-end video games, where I don’t need to display as many textures at once. As I need a lot of video memory, it can be difficult to find the right balance, but I will always buy the best GPU I can afford to help boost my workflow.


How a diverse industry creates diverse thinking


3D artists come into the field from diverse backgrounds, both technical and artistic. To be a 3D artist, you need some creative skill, but creativity is something you can nurture throughout your career. As for technical skills, if you don’t already possess these you can learn (although it’s harder if you don’t have a background in computing). Saying that, today’s software tools are far more accessible and user-friendly, helped by powerful hardware that speeds up the workflow and improves performance. For instance, nowadays you don’t have to start modelling with a base mesh, you can create forms and shapes without using any mathematical or technical approach. So effectively whatever jumps out of your head can materialise as a 3D model.



While technical skills and an artistic background will make your first steps in your career as a 3D artist easier, patience is also a key skill, as you’ll need to work for many hours a day to hone your craft. I recently taught two people from scratch, neither of whom had much experience with specialised computer software, and now they are working as successful professional 3D artists. Both have a high level of patience, persistence and a willingness to learn and develop.


Being Polish, it’s also exciting to see the growth of this industry in my home country. We have great 3D artists and developers here, and with today’s modern technology, we are now afforded the same level of creative freedom as those users in western Europe and America, where the biggest productions are made.


Our home-grown game development industry is evolving, with studios including Flying Wild Hog, The Astronauts, 11 Bit Studios, CI Games, Techland, People Can Fly and CD PROJEKT RED – of course, the much-anticipated Cyberpunk 2077 is on the way next year. Not only are these companies attracting employees from around the world but investors too, drawn by the investment opportunities in our local industry.



AR & VR: The next frontier


While powerful hardware is currently used to provide the best possible experience to users via a screen in front of them – through the use of, for example, AMD FreeSync Technology and 8K screens – the next frontier will be immersive experiences, such as VR and AR.



The professional applications of VR and AR is already advancing, with 3D artists using headsets such as Oculus and HTC Vive to sculpt and draw in VR. This will be the next step on our collective technology journey and an extremely exciting one for creatives here in Europe and beyond.




About Pawel Jaruga


Pawel "Levus3D" Jaruga is a character artist, digital sculptor and instructor based in Poland. He has over 10 years’ experience in games, commercials and cinematics industry. He’s also the owner of, miniatures and collectibles studio. You can view his work here.


Notable game credits:

  • Witchfire (The Astronauts, TBC)
  • Hard Reset: Redux (Flying Wild Hog, 2016)
  • Shadow Warrior 2 (Flying Wild Hog, 2016)
  • Ryse: Son of Rome - Legendary Edition (Crytek/Microsoft, 2014)
  • Ryse: Son of Rome (Crytek/Microsoft, 2013)
  • Shadow Warrior (Flying Wild Hog, 2013)
  • Hard Reset: Exile (Flying Wild Hog, 2012)
  • Hard Reset (Flying Wild Hog, 2011)
  • Ancient Quest of Saqqarah (Codeminion, 2008)
  • Stoneloops! of Jurassica (Codeminion, 2008)


Specific focus areas:

  • 3D Design, Modelling and Printing
  • Character Design and Modelling
  • Creature Design and Modelling
  • Digital Sculpting
  • Physically Based Rendering
  • Texturing


Software used:

  • AMD ProRender
  • Zbrush
  • 3DCoat
  • 3DS Max
  • Blender
  • Marmoset
  • Marvelous Designer
  • Octane Render
  • Substance Painter


Pawel Jaruga received a Radeon Pro WX9100 graphics card in exchange for his blog contribution. The blog represents Mr. Jaruga’s own thoughts and opinions as of the date published. AMD and/or the third-party blogger have no obligation to update any forward-looking content in the above blog. AMD is not responsible for the content of any third-party and does not necessarily endorse the comments made therein. 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.

Today I am excited to let you know that the AMD EPYC Cloud footprint is increasing globally with Tencent Cloud announcing its 2nd Gen AMD EPYC processor-based “Star Lake” Server Platform at the Tencent Global Digital Ecosystem Conference.


Let’s take a closer look at the latest Tencent Cloud announcement and how the 2nd Gen AMD EPYC processor-powered “Star Lake” platform enables Tencent Cloud to achieve their business goals and extend performance & TCO advantages to their customers.


Enhanced efficiency and scalable performance for exponentially growing cloud service

Tencent recently became the first Chinese company with more than 1.1 million servers in their network and is one of the fastest growing cloud computing companies in the global IaaS market. This rapid business growth brings new challenges including efficiency improvement and operational cost reduction for the infrastructure. To address these challenges Tencent Cloud developed a technology system from the infrastructure layer to the application layer to enable the next stage of cloud computing growth.


Server design, energy efficiency, security features, and reliability have a direct impact on the performance and cost efficiency of Cloud Service Provider offerings. The 2nd Gen AMD EPYC processor based “Star Lake” server platform is Tencent’s first self-designed server developed for the Tencent Cloud environment. Tencent employed many advanced capabilities to improve energy efficiency. For example, according to Tencent, the advanced thermosyphon heat dissipation technology used in “Star Lake” improves maximum load energy efficiency by 50%. The “Star Lake” platform is designed to optimize cloud computing, storage and network requirements to effectively meet 98% of Tencent cloud application scenarios.


Liu YuXun, General Manager of Tencent's server supply chain Announcing the AMD EPYC<img src='' class='jive_emoji'/> Powered Star Lake Server Platform at the Tencent Global Digital Ecosystem Conference, 2019.

Liu YuXun, General Manager of Tencent's server supply chain Announcing the AMD EPYC Powered Star Lake Server Platform at the Tencent Global Digital Ecosystem Conference, 2019.


Industry's best single core performance and significant single core TCO savings with 2nd Gen AMD EPYC Processor Powered “Star Lake” server

According to test results presented by Tencent Cloud at the Tencent Global Digital Ecosystem Conference, the “Star Lake” Server with Tencent’s SA 2 instance powered by 2nd Gen AMD EPYC processors achieved the industry’s best single core performance and provides a significant TCO advantage. Tencent Cloud’s results in the image above show that the 2nd Gen AMD EPYC processor-based “Star Lake” server platform improves the overall performance of Tencent’s SA 2 cloud service instances by 35%, including 40% improvement in video processing, 35% improvement in graphics transcoding and 150% improvement in page QPS. This enables Tencent Cloud to provide performance enhancements and cost efficiencies to end customers.


You can read more about this at the Tencent Cloud Star Lake Announcement here. It's in Mandarin but can be translated easily!


I greatly appreciate the close collaboration and efforts of the Tencent Cloud and AMD teams to bring these innovations to our customers.

We at AMD are proud to be at the forefront of innovation through our collaboration with Microsoft Azure to offer our latest innovations to cloud-based enterprises with the general availability of the new Azure D-series and E-series virtual machines powered by AMD EPYC 7452 Processors. AMD and Microsoft Azure will continue our collaboration to provide guidance on optimization & migration to Azure virtual machines powered by AMD EPYC Processors. AMD and Microsoft are also expanding their partnership with Azure Data Explorer, a leading managed data analytics service for near real-time ingestion and ultra-fast queries.


Operational cost efficiency, space optimization, and faster application response times are critical for today’s modern data centers. Architectural innovations in AMD EPYC 7002 Series processors are designed to deliver exceptional performance and scalability to help drive TCO savings for users of a variety of cloud environments including traditional bare metal, software defined, converged and hyper-converged infrastructures in private, public, and hybrid cloud environments.


Let’s take a quick look at how Azure and AMD EPYC continue to give customers leadership performance for cloud workloads.


Enhanced Performance with Azure D-series virtual machines powered by AMD EPYC

Microsoft considers the Azure Da_v4 and Das_v4-Series the fastest Azure VMs in their class, with a balanced core-to-memory ratio, providing enhanced performance for a wide variety of production workloads. Example use cases include most enterprise-grade applications, relational databases, in-memory caching, and analytics. Microsoft Azure D-series virtual machines are powered by AMD EPYC 7452 Processors and provide up to 96 vCPUs, 384GB DDR4 RAM, and 2.4TB of SSD-based temporary storage per virtual machine.


Optimize large in-memory business critical workloads with Microsoft Azure E-Series virtual machines powered by AMD EPYC

Azure Ea_v4 and Eas_v4 VMs offer class-leading performance for memory-intensive applications such as relational databases, caching servers, and in-memory analytics. Powered by AMD EPYC 7452 Processors, the E-Series offer up to 96 vCPUs, up to 672GB DDR4 memory, and 2.4TB SSD-based temporary storage per VM. For database workloads, the Ea-series VMs offers a 22% better performance/dollar than competitive VMs. [i]


Power a lightning fast data exploration engine

AMD and Microsoft are expanding their partnership with Azure Data Explorer, a leading managed data analytics service for near real-time ingestion and ultra-fast queries. Azure Data Explorer is using commercially available Azure compute powered by AMD EPYC to deliver groundbreaking and cost-effective interactive analytics.


Microsoft Ignite offers a great opportunity to explore innovative ways to build solutions, migrate and manage your infrastructure, using the new Azure D-series and E-series virtual machines powered by AMD EPYC processors.


In addition, there are plenty of chances to learning the latest skills from technology leaders and industry users shaping the future of cloud. AMD is hosting a technical breakout session (BRK1114: “Turbocharge your infrastructure with AMD EPYC”) on Thursday, November 7 at 11:30AM-12:15PM in OCCC W208. You can also come by meeting room MR-32, Sponsor Rooms B in the Partner Solution Zone for a deeper dive into our innovative technologies or join us at Booth # 249 to experience solution demos and interact with AMD experts.


You can also read more about the new Azure VMs on the Microsoft blog, here.


I would like to thank the Microsoft and AMD teams who partnered to bring these innovations to our customers.


[i] "Results as of 10-28-2019 using MS SQL Server 2019. Comparison based on internal testing of HammerDB TPCC/OLTP workload. Azure E16asv4 virtual machine generated a result of 600K transactions/minute and costs $0.5301/hour based on three year reserve pricing in US East with RHEL operating system. Pricing found at AWS r5.4xlarge virtual machine generated a result of 545K transactions/minute and costs $0.587/hour based on effective hourly 3-year reserve pricing in US East region with RHEL operating system. Pricing found at ROM-340