The Channel Company

Sponsored by:

Visit AMD Visit Supermicro

Performance Intensive Computing

Capture the full potential of IT

How Ahrefs speeds SEO services with huge compute, memory & storage

Featured content

How Ahrefs speeds SEO services with huge compute, memory & storage

Ahrefs, a supplier of search engine optimization tools, needed more robust tech to serve its tens of thousands of customers and crawl billions of web pages daily. The solution: More than 600 Supermicro Hyper servers powered by AMD processors and loaded with huge memory and storage.

Learn More about this topic
  • Applications:
  • Featured Technologies:

Wondering how to satisfy customers who need big—really big—compute and storage? Take a tip from Ahrefs Ltd.

This company, based in Singapore, is a 10-year-old provider of search engine optimization (SEO) tools.

Ahrefs has a web crawler that processes up to 8 billion pages a day. That makes Ahrefs one of the world’s biggest web crawlers, up there with Google and Bing, according to internet hub Cloudflare Radar.

What’s more, Ahrefs’ business has been booming. The company now has tens of thousands of users.

That’s good news. But it also meant that to serve these customers, Ahrefs needed more compute power and storage capacity. And not just a little more. A lot.

Ahrefs also realized that its current generation of servers and CPUs couldn’t meet this rising demand. Instead, the company needed something new and more powerful.

Gearing up

For Ahrefs, that something new is its recent order of more than 600 Supermicro servers. Each system is equipped with dual      4th generation AMD EPYC 9004 Series processor, a whopping 1.5 TB of DDR5 memory, and a massive 120+ TB of storage.

More specifically, Ahrefs selected Supermicro’s AS-2125HS-TNR servers. They’re powered by dual AMD EPYC 9554 processors, each with 64 cores and 128 threads, running at a base clock speed of 3.1 GHz and an all-core boost speed of 3.75 GHz.

For Ahrefs’ configuration, each Supermicro server also contains eight NVMe 15.3 TB SSD storage devices, for a storage total of 122 TB. Also, each server communicates with the Ahrefs data network via two 100 Gbps ports.

Did it work?

Yes. Ahrefs’ response times got faster, even as its volume increased. The company can now offer more services to more customers. And that means more revenue.

Ahrefs’ founder and CEO, Dimitry Gerasimenko, puts it this way: “Supermicro’s AMD-based servers were an ideal fit for our business.”

How about you? Have customers who need really big compute and storage? Tell them about Ahrefs, and point them to these resources:

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

Do you know why 64 cores really matters?

Featured content

Do you know why 64 cores really matters?

In a recent test, Supermicro workstations and servers powered by 3rd gen AMD Ryzen Threadripper PRO processors ran engineering simulations nearly as fast as a dual-processor system, but needed only two-thirds as much power.

Learn More about this topic
  • Applications:
  • Featured Technologies:

More cores per CPU sounds good, but what does it actually mean for your customers?

In the case of certain Supermicro workstations and servers powered by 3rd gen AMD Ryzen Threadripper PRO processors, it means running engineering simulations with dual-processor performance from a single-socket system. And with further cost savings due to two-thirds lower power consumption.

That’s according to tests recently conducted by MVConcept, a consulting firm that provides hardware and software optimizations. The firm tested two Supermicro systems, the AS-5014A-TT SuperWorkstation and AS-2114GT-DPNR server.

A solution brief based on MVConcept’s testing is now available from Supermicro.

Test setup

For these tests, the Supermicro server and workstation were both tested in two AMD configurations:

  • One with the AMD Ryzen Threadripper PRO 5995WX processor
  • The other with an older, 2nd gen AMD Ryzen Threadripper PRO 3995WX processor

In the tests, both AMD processors were used to run 32-core as well as 64-core operations.

The Supermicro systems were tested running Ansys Fluent, fluid simulation software from Ansys Inc. Fluent models fluid flow, heat, mass transfer and chemical reactions. Benchmarks for the testing included aircraft wing, oil rig and pump.

The results

Among the results: The Supermicro systems delivered nearly dual-CPU performance with a single processor, while also consuming less electricity.

What’s more, the 3rd generation AMD 5995WX CPU delivered significantly better performance than the 2nd generation AMD 3995WX.

Systems with larger cache saw performance improved the most. So a system with L3 cache of 256MB outperformed one with just 128MB.

BIOS settings proved to be especially important for realizing the optimal performance from the AMD Ryzen Threadripper PRO when running the tested applications. Specifically, Supermicro recommends using NPS=4 and SMT=OFF when running Ansys Fluent with AMD Ryzen Threadripper PRO. (NPS = non-uniform memory access (NUMA) per socket; and SMT = symmetric multithreading.)

Another cool factor involves taking advantage of the Supermicro AS-2114GT-DPNR server’s two hot-pluggable nodes. First, one node can be used to pre-process the data. Then the other node can be used to run Ansys Fluid.

Put it all together, and you get a powerful takeaway for your customers: These AMD-powered Supermicro systems offer data-center power on both the desktop and server rack, making them ideal for SMBs and enterprises alike.

Do more:

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

Try before you buy with Supermicro’s H13 JumpStart remote access program

Featured content

Try before you buy with Supermicro’s H13 JumpStart remote access program

The Supermicro H13 JumpStart Remote Access program lets you and your customers test data-center workloads on Supermicro systems based on 4th Gen AMD EPYC 9004 Series processors. Even better, the program is free.

Learn More about this topic
  • Applications:
  • Featured Technologies:

You and your customers can now try out systems based on 4th Gen AMD EPYC 9004 Series processors at no cost with the Supermicro remote access program.

Called H13 JumpStart, the free program offers remote access to Supermicro’s top-end H13 systems.

Supermicro’s H13 systems are designed for today’s advanced data-center workloads. They feature 4th Gen AMD EPYC 9004 Series processors with up to 96 Zen 4 cores per socket, DDR5 memory, PCIe 5.0, and support for Compute Express Link (CXL) 1.1+ peripherals.

The H13 JumpStart program lets you and your customers validate, test and benchmark workloads on either of two Supermicro systems:

●      Hyper AS-2025HS-TNR: Features dual AMD EPYC processors, 24 DIMMS, up to 3 accelerator cards, AIOM network adapter, and 12 hot-swap NVMe/SAS/SATA drive bays.

●      CloudDC AS-2015CS-TNR: Features a single AMD processor, 12 DIMMS, 4 accelerator cards, dual AIOM network adapters, and a 240GB solid state drive.

Simple startup

Getting started with Supermicro’s H13 JumpStart program is simple. Just sign up with your name, email and a brief description of what you plan to do with the system.

Next, Supermicro will verify your information and your request. Assuming you qualify, you’ll receive a welcome email from Supermicro, and you’ll be scheduled to gain access to the JumpStart server.

Next, you’ll be given a unique username, password and URL to access your JumpStart account.

Run your test. Once you’re done, Supermicro will also ask you to complete a quick survey for your feedback on the program.

Other details

The JumpStart program does have a few limitations. One is the number of sessions you can have open at once. Currently, it’s limited to 1 VNC (virtual network computing), 1 SSH (secure shell), and 1 IPMI (intelligent platform management interface) session per user.

Also, the JumpStart test server is not directly addressable to the internet. However, the servers can reach out to the internet to get files.

You should test with JumpStart using anonymized data only. That’s because the Supermicro server’s security policies may differ from those of your organization.

But rest assured, once you’re done with your JumpStart demo, the server storage is manually erased, the BIOS and firmware are reflashed, and the OS is re-installed with new credentials. So your data and personal information are completely removed.

Get started

Ready to get a jump-start with Supermicro’s H13 JumpStart Remote Access program? Apply now to secure access.

Want to learn more about Supermicro’s H13 system portfolio? Check out a 5-part video series featuring Linus Sebastian of Linus Tech Tips. He takes a deep dive into how these Supermicro systems run faster and greener. 

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

How rackscale integration can help your customers get productive faster

Featured content

How rackscale integration can help your customers get productive faster

Supermicro’s rack integration and deployment service can help your customers get productive sooner.

 

Learn More about this topic
  • Applications:
  • Featured Technologies:

How would your key data-center customers like to improve their server performance, speed their rate of innovation, and lower their organization’s environmental impact—all while getting productive sooner?

Those are among the key benefits of Supermicro’s rack integration and deployment service. It’s basically a one-stop shop for a defined process with experts to design and create an effective and efficient cloud and enterprise hardware solution.

Supermicro’s dedicated team can provide everything from early design to onsite integration. That includes design, assembly, configuration, testing and delivery.

Hardware covered by Supermicro’s rack integration service includes servers, storage, switches and rack products. That includes systems based on the latest 4th Generation AMD EPYC server processors. Supermicro’s experts can also work closely with your customer to design a test plan that includes application loading, performance tuning and testing.

All these can be used for a wide range of optimized solutions. These include AI and deep learning, big data and Hadoop refreshes, and vSAN.

Customers of Supermicro’s rackscale systems can also opt for liquid cooling. This can reduce your customer’s operating expenses by more than 40%. And by lowering fan speeds, liquid cooling can further reduce their power needs, delivering a PUE (power usage effectiveness metric) of close to 1.0. All that typically provides an ROI in just 1 year, according to Supermicro.

Five-phase integration

When your customers work with Supermicro on rack integration, they’ll get support through 5 phases:

  • Design: Supermicro learns your customer’s business problems and requirements, develops a proof-of-concept to validate the solution, then selects the most suitable hardware and works with your customer on power requirements and budgets. Then it creates a bill of materials, followed by a detailed rack-level engineering diagram.
  • Assembly: Supermicro technicians familiar with the company’s servers assemble the system, either on your customer’s site or pre-shipment at a Supermicro facility. This includes all nodes, racks, cabling and third-party equipment.
  • Configuration: Each server’s BIOS is updated, optimized and tested. Firmware gets updated, too. OSes and custom images are pre-installed or deployed to specific nodes as needed.
  • Testing: This includes a performance analysis, a check for multi-vendor compatibility, and full rack burn-in testing for a standard 8 hours.
  • Logistics: Supermicro ships the complete system to your customer’s site, can install it, and provides ongoing customer service.

Big benes

For your customers, the benefits of working with Supermicro and AMD can include better performance per watt and per dollar, faster time to market with IT innovation, a reduced environmental impact, and lower costs.

Further, once the system is installed, Supermicro’s support can significantly reduce lead times to fix system issues. The company keeps the whole process from L6 to L12 in-house, and it maintains a vast inventory of spare parts on campus.

Wherever your customers are located, Supermicro likely has an office nearby. With a global footprint, Supermicro operates across the U.S., EMEA and Taiwan. Supermicro has invested heavily in rack-integration testing facilities, too. These centers are now being expanded to test rack-level air and liquid cooling.

For your customers with cloud-based systems, there are additional benefits. These include optimizing the IT environment for their clouds, and meeting co-location requirements.

There’s business for channel partners, too. You can add specific software to the rack system. And you can work with your customer on training and more.

Do more:

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

AMD-based servers support enterprise applications — and break OLTP records

Featured content

AMD-based servers support enterprise applications — and break OLTP records

AMD EPYC server processors are designed to help your data-center customers get their workloads done faster and with fewer computing resources.

 

Learn More about this topic
  • Applications:
  • Featured Technologies:

AMD EPYC™ server processors are designed to help your data-center customers get their workloads done faster and with fewer computing resources.

AMD EPYC server processors offer a consistent set of features across a range of choices from 8 to 96 cores. This balanced set of resources found in AMD EPYC processors lets your customers right-size server configurations to fit their workloads.

What’s more, these AMD CPUs include models that offer high per-core performance optimized for frequency-sensitive and single-threaded workloads. This can help reduce the TCO for core-based software licenses.

AMD introduced the 4th Generation AMD EPYC processors in late 2022. The first of this generation are the AMD EPYC 9004 series CPUs. They’ve been designed to support performance and efficiency, help keep data secure, and use the latest industry features and architectures.

AMD continues to ship and support the previous 3rd Generation AMD EPYC 7002 and 7003 series processors. These processors power servers that are now available from a long list of leading hardware suppliers, including Supermicro.

Record-breaking

Good as all that may sound, you and your customers still need hard evidence that AMD processors can truly speed up their enterprise applications. Well, a new independent test of AMD-based Supermicro servers has provided just that.

The test was performed by the Telecommunications Technology Association (TTA), an IT standardization association based in Seongnam, South Korea. The TTA tested several Supermicro database and web servers powered by 3rd Gen AMD EPYC 7343 processors.

The results: The Supermicro servers set a world record for performance by a non-cluster system of 507,802 transactions per minute (tpmC).

That test was conducted using the TPC Benchmark, which measures a server’s online transaction processing (OLTP) performance. The tpmC metric measures how many new-order transactions a system can generate in a minute while executing business transactions under specific response-time requirements.

What’s more, when compared with servers based on the previous 2nd Gen AMD EPYC processors, the newer Supermicro servers were 33% faster, as shown in the chart below:

DATA: Telecommunications Technology Association

All that leads the TTA to conclude that Supermicro servers powered by the latest AMD processors “empower organizations to create deployments that deliver data insights faster than ever before.”

Do more:

Note:

1. https://www.tpc.org/1809

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

For Greener Data Centers, Look to Energy-Efficient Components

Featured content

For Greener Data Centers, Look to Energy-Efficient Components

Energy-efficient systems can help your customers lower their data-center costs while supporting a cleaner environment. 

Learn More about this topic
  • Applications:
  • Featured Technologies:

Creating a more energy-efficient data center isn’t only good for the environment, but also a great way for your customers to lower their total cost of ownership (TCO).

In many organizations, the IT department is the single biggest consumer of power. Data centers are filled with power-hungry components, including servers, storage devices, air conditioning and cooling systems.

The average data center uses anywhere from 2 to 4 Terawatt hours (TWh) of electricity per year. That works out to nearly 3% of total global energy use, according to Supermicro. Looking ahead, that’s forecast to reach as high as 8% by 2030.

One important measure of data-center efficiency is Power Usage Effectiveness (PUE). It’s calculated by taking the total electricity in a data center and dividing it by the electricity used by center’s IT components. The difference is how much electricity is being used for cooling, lighting and other non-IT components.

The lower a data center’s PUE, the better. The most energy-efficient data centers have a PUE of 1.0 or lower. The average PUE worldwide last year was 1.55, says the Uptime Institute, a benchmarking organization. That marked a slight improvement over 2021, when the average PUE was 1.57.

Costly power

All that power is expensive, too. Among the short list of ways your customers can lower that cost, moving to energy-efficient server CPUs is especially effective.

For example, AMD says that 11 servers based on of its 4th gen AMD EPYC processors can use up to 29% less power a year than the 17 servers based on competitive CPUs required to handle the same workload volume. And that can help reduce an organization’s capital expenditures by up to 46%, according to AMD.

As that example shows, CPUs with more cores can also reduce power needs by handling the same workloads with fewer physical servers.

Yes, a high-core CPU typically consumes more power than one with fewer cores, especially when run at the same frequency. But by handling more workload volume, a high-core CPU lets your customer do the same or more work with fewer racks. That can also reduce the real estate footprint and lower the need for cooling.

Greener tactics

Other tactics can contribute to a greener data center, too.

One approach involves what Supermicro calls a “disaggregated” server architecture. Essentially, this means that a server’s subsystems—including its CPU, memory and storage—can be upgraded without having to replace the entire chassis. For a double benefit, this lowers TCO while reducing E-waste.

Another approach involves designing servers that can share certain resources, such as power supplies and fans. This can lower power needs by up to 10%, Supermicro says.

Yet another approach is designing servers for maximum airflow, another Supermicro feature. This allows the CPU to operate at higher temperatures, reducing the need for air cooling.

It can also lower the load on a server’s fans. That’s a big deal, because a server’s fans can consume up to 15% of its total power.

Supermicro is also designing systems for liquid cooling. This allows a server’s fan to run at a lower speed, reducing its power needs. Liquid cooling can also lower the need for air conditioning, which in turn lowers PUE.

Liquid cooling functions similarly to a car’s radiator system. It’s basically a circular system involving an external “chiller” that cools the liquid and a series of pipes. The liquid is pumped to run through one or more pipes over a server’s CPU and GPU. The heat from those components warms the liquid. Then the now-hot liquid is sent back to the chiller for cooling and then recirculation.

Green vendors

Leading suppliers can help you help your customers go green.

AMD, for one, has pledged itself to delivering a 30x increase in energy efficiency for its processors and accelerators by 2025. That should translate into a 97% reduction in energy use per computation.

Similarly, Supermicro is working hard to help customers create green data centers. The company participates in industry consortia focused on new cooling alternatives and is a leader in the Liquid Cooling Standing Working Group of The Green Grid, a membership organization that fosters energy-efficient data centers.

Supermicro also offers products using its disaggregated rack-scale design approach to offer higher efficiency and lower costs.

Do more:

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

Learn, Earn and Win with AMD Arena

Featured content

Learn, Earn and Win with AMD Arena

Channel partners can learn about AMD products and technologies at the AMD Arena site. It’s your site for AMD partner training courses, redeemable points and much more.

Learn More about this topic
  • Applications:
  • Featured Technologies:

Interested in learning more about AMD products while also earning points you can redeem for valuable merch? Then check out the AMD Arena site.

There, you can:

  • Stay current on the latest AMD products with training courses, sales tools, webinars and quizzes;
  • Earn points, unlock levels and secure your place in the leaderboard;
  • Redeem those points for valuable products, experiences and merchandise in the AMD Rewards store.

Registering for AMD Arena is quick, easy and free. Once you’re in, you’ll have an Arena Dashboard as your control center. It’s where you can control your profile, begin a mission, track your progress, and view your collection of badges.

Missions are made of learning objectives that take you through training courses, sales tools, webinars and quizzes. Complete a mission, and you can earn points, badges and chips; unlock levels; and climb the leaderboard.

The more missions you complete, the more rewards you’ll earn. These include points you can redeem for merchandise, experiences and more from the AMD Arena Rewards Store.

Courses galore

Training courses are at the heart of the AMD Arena site. Here are just 3 of the many training courses waiting for you now:

  • AMD EPYC Processor Tool: Leverage the AMD processor-selector and total cost of ownership (TCO) tools to match your customers’ needs with the right AMD EPYC processor.
  • AMD EPYC Processor – Myth Busters: Get help fighting the myths and misconceptions around these powerful CPUs. Then show your data-center customers the way AMD EPYC delivers performance, security and scalability.

Get started

There’s lots more training in AMD Arena, too. The site supports virtually all AMD products across all business segments. So you can learn about both products you already sell as well as new products you’d like to cross-sell in the future.

To learn more, you can take this short training course: Introducing AMD Arena. In just 10 minutes, this course covers how to register for an AMD Arena account, use the Dashboard, complete missions and earn rewards.

Ready to learn, earn and win with AMD Arena? Visit AMD Arena now. 

 

 

Featured videos

Events

Find AMD & Supermicro Elsewhere

Supermicro H13 Servers Maximize Your High-Performance Data Center

Featured content

Supermicro H13 Servers Maximize Your High-Performance Data Center

Learn More about this topic
  • Applications:
  • Featured Technologies:
  • Featured Companies:
  • AMD

The modern data center must be both highly performant and energy efficient. Massive amounts of data are generated at the edge and then analyzed in the data center. New CPU technologies are constantly being developed that can analyze data, determine the best course of action, and speed up the time to understand the world around us and make better decisions.

With the digital transformation continuing, a wide range of data acquisition, storage and computing systems continue to evolve with each generation of  a CPU. The latest CPU generations continue to innovate within their core computational units and in the technology to communicate with memory, storage devices, networking and accelerators.

Servers and, by default, the CPUs within those servers, form a continuum of computing and I/O power. The combination of cores, clock rates, memory access, path width and performance contribute to specific servers for workloads. In addition, the server that houses the CPUs may take different form factors and be used when the environment where the server is placed has airflow or power restrictions. The key for a server manufacturer to be able to address a wide range of applications is to use a building block approach to designing new systems. In this way, a range of systems can be simultaneously released in many form factors, each tailored to the operating environment.

The new H13 Supermicro product line, based on 4th Generation AMD EPYC™ CPUs, supports a broad spectrum of workloads and excels at helping a business achieve its goals.

Get speeds, feeds and other specs on Supermicro’s latest line-up of servers

Featured videos

Events

Find AMD & Supermicro Elsewhere

Perspective: Looking Back on the Rise of Supercomputing

Featured content

Perspective: Looking Back on the Rise of Supercomputing

Learn More about this topic
  • Applications:
  • Featured Technologies:

We’ve come a long way on the development of high performance computing. Back in 2004, I attended an event held in the gym at the University of San Francisco. The goal was to crowd-source computing power by connecting the PCs of volunteers who were participating in the first “Flash Mob Computing” cluster computing event. Several hundred PCs were networked together in the hope that they would create one of the largest supercomputers, albeit for a few hours.

 

I brought two laptops for the cause. The participation rules stated that the  data on our hard drives would remain intact. Each computer would run a specially crafted boot CD that ran a benchmark called Linpack, a software library for performing numerical linear algebra running on Linux. It was used to measure the collective computing power.

 

The event attracted people with water-cooled overclocked PCs, naked PCs (no cases, just the boards and other components) and custom-made rigs with fancy cases. After a few hours, we had roughly 650 PCs on the floor of the gym. Each PC was connected to a bunch of Foundry BigIron super-switches that were located around the room.

 

The 2004 experiment brought out several industry luminaries, such as Gordon Bell, who was the father of the Digital Equipment Corporation VAX minicomputer, and Jim Gray, who was one of the original designers behind the TPC benchmark while he was at Tandem. Both men at the time were Microsoft fellows. Bell was carrying his own laptop but had forgotten to bring his CD drive, so he couldn’t connect to the mob.

 

Network shortcomings

 

What was most interesting to me, and what gave rise to the mob’s eventual undoing, were the networking issues involved with assembling and running such a huge collection of gear. The mob used ordinary 100BaseT Ethernet, which was a double-edged sword. While easy to set up, it was difficult to debug when network problems arose. The Linpack benchmark requires all the component machines to be running concurrently during the test, and the organizers had trouble getting all 600-plus PCs to operate online flawlessly. The best benchmark accomplished was a peak rate of 180 gigaflops using 256 computers, but that wasn’t an official score as one node failed during the test.

 

To give you an idea of where this stood in terms of overall supercomputing prowess, it was better than the Cray supercomputers of the early 1990s, which delivered around 16 gigaflops.If you lo

 

At the website top500.org (which tracks the fastest supercomputers around the globe), you can see that all the current top 500 machines are measured in petaflops (1 million gigaflops). The Oak Ridge National Laboratory’s Frontier machine, which has occupied the number one spot this year, weighs in at more than 1,000 petaflops and uses 8 million cores. To make the fastest 500 list back in 2004, the mob would have had to achieve a benchmark of over 600 gigaflops. Because of the networking problems, we’ll never know for sure.Still, it was an impressive achievement, given the motley mix of machines. All of the world’s top 500 supercomputers are custom built and carefully curated and assembled to attain that level of computing performance.

 

Another historical note: back in 2004, one of the more interesting entries came in third on the top500.org list: a collection of several thousand Apple Macintoshes running at Virginia Polytechnic University. Back in the present, as you might imagine, almost all the fastest 500 supercomputers are based on a combination of CPU and GPU chip architectures.

 

Today, you can buy your own supercomputer on the retail market, such as the Supermicro SuperBlade® models. And of course, you can routinely run much faster networking protocols than 100-megabit Ethernet.

Featured videos

Events

Find AMD & Supermicro Elsewhere

Supermicro SuperBlades®: Designed to Power Through Distributed AI/ML Training Models

Featured content

Supermicro SuperBlades®: Designed to Power Through Distributed AI/ML Training Models

Running heavy AI/ML workloads can be a challenge for any server, but the SuperBlade has extremely fast networking options, upgradability, the ability to run two AMD EPYC™ 7000-series 64-core processors and the Horovod open-source framework for scaling deep-learning training across multiple GPUs.

Learn More about this topic
  • Applications:
  • Featured Technologies:

Running the largest artificial intelligence (AI) and machine learning (ML) workloads is a job for the higher-performing systems. Such loads are often tough for even more capable machines. Supermicro’s SuperBlade combines blades using AMD EPYC™ CPUs with competing GPUs into a single rack-mounted enclosure (such as the Supermicro SBE-820H-822). That leverages an extremely fast networking architecture for these demanding applications that need to communicate with other servers to complete a task.

 

The Supermicro SuperBlade fits everything into an 8U chassis that can host up to 20 individual servers. This means a single chassis can be divided into separate training and model processing jobs. The components are key: servers can take advantage of the 200G HDR InfiniBand network switch without losing any performance. Think of this as delivering a cloud-in-a-box, providing both easier management of the cluster along with higher performance and lower latencies.

 

The Supermicro SuperBlade is also designed as a disaggregated server, meaning that components can be upgraded with newer and more efficient CPUs or memory as technology progresses. This feature significantly reduces E-waste.


The SuperBlade line supports a wide selection of various configurations, including both CPU-only and mixed CPU/GPU models, such as the SBA-4119SG, which comes with up to two AMD EPYC™ 7000-series 64-core CPUs. These components are delivered on blades that can easily slide right in. Plus, they slide out as easily when you need to replace the blades or the enclosure. The SuperBlade servers support a wide network selection as well, ranging from 10G to 200G Ethernet connections.

 

The SuperBlade employs the Horovod distributed model-training, message-passing interface to let multiple ML sessions run in parallel, maximizing performance. In a sample test of two SuperBlade nodes, the solution was able to process 3,622 GoogleNet images/second, and eight nodes were able to scale up to 13,475 GoogleNet images/second.


As you can see, Supermicro’s SuperBlade improves performance-intensive computing and boosts AI and ML use cases, enabling larger models and data workloads. The combined solution enables higher operational efficiency to automatically streamline processes, monitor for potential breakdowns, apply fixes, more efficiently facilitate the flow of accurate and actionable data and scale up training across multiple nodes.

Featured videos

Events

Find AMD & Supermicro Elsewhere

Pages