The Channel Company

Sponsored by:

Visit AMD Visit Supermicro

Performance Intensive Computing

Capture the full potential of IT

AMD intros entry-level server CPUs for SMBs, enterprise branches, hosted service providers

Featured content

AMD intros entry-level server CPUs for SMBs, enterprise branches, hosted service providers

The new AMD EPYC 4004 processors extend the company’s ‘Zen 4’ core architecture into a line of entry-level systems for small and midsized businesses, schools, branch IT and regional providers of hosted IT services.

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

AMD has just introduced the AMD EPYC 4004 processors, bringing a new entry-level line to its family of 4th gen server processors.

To deliver these new processors, AMD has combined the architecture of its Ryzen 7000 series processors with the packaging of its EPYC line of server processors. The result is a line of CPUs that lowers the entry-level pricing for EPYC-powered servers.

The AMD EPYC 4004 processors are designed for use in entry-level servers and towers, systems that typically retail for $1,500 to $3,000. That’s a price level affordable for most small and medium businesses, enterprise IT branches, public school districts, and regional providers of hosted IT services. It’s even less than the retail price for some high-end processor CPUs.

Many SMBs can’t afford either hosting on the public cloud or AMD’s more powerful server processors. As a result, they often make do with using PCs as servers. The new AMD processors aim to change that.

There are lots of reasons why a real server offers a better solution. These reasons include greater performance and scalability, higher rates of dependability and easier management.

Under the hood

The new AMD EPYC 4004 series is initially offered as eight SKUs, all designed for use in single-processor systems. They offer from 8 to 16 ‘Zen 4’ cores with up to 32 threads; 128MB of L3 cache; 2 DDR channels with a memory capacity of up to 192GB; and 28 lanes of PCIe 5 connectivity.

Two of the new SKUs—4584PX and 4484PX—offer AMD’s 128MB 3D V-Cache technology. As the name implies, V-Cache is a 3D vertical cache designed to offer faster interconnect density, greater energy efficiency and higher per-core performance for cache-hungry applications.

All the new AMD EPYC 4004 processors use AMD’s AM5 socket. That makes them incompatible with AMD’s higher-end EPYC 8004 and EPYC 9004 server processors, which use a different socket.

OEM support

AMD is working with several server OEMs to get systems built around the new EPYC 4004 processors to market quickly. Among these OEMs is Supermicro, which is supporting the new AMD CPUs in select towers and servers.

That includes Supermicro’s H13 MicroCloud system, a high-density, 3U rackmount system for the cloud. It has now been updated with additional performance offered by the AMD EPYC 4004.

Supermicro’s H13 MicroCloud retails for about $10K, making it more expensive than most entry-level servers. But unlike those less-expensive servers, the MicroCloud offers 8 single-processor nodes for applications requiring multiple discrete servers, such as e-commerce sites, code development, cloud gaming and content creation.

AMD says shipments of the new AMD EPYC 4004 Series processors, as well as of OEM systems powered by the new CPUs, are expected to begin during the first week of June. Pre-sales orders of the new processors, AMD adds, have already been strong.

Do more:

 

Featured videos

Follow

Related Content

Tech Explainer: Why the Rack is Now the Unit

Featured content

Tech Explainer: Why the Rack is Now the Unit

Today’s rack scale solutions can include just about any standard data center component. They can also save your customers money, time and manpower.

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

Are your data center customers still installing single servers and storage devices instead of full-rack solutions? If so, they need to step up their game. Today, IT infrastructure management is shifting toward rack scale integrations. Increasingly, the rack is the unit.

A rack scale solution can include just about any standard data center component. A typical build combines servers, storage devices, network switches and other rack products like power-management and cooling systems. Some racks are loaded with the same type of servers, making optimization and maintenance easier.

With many organizations developing and deploying resource-intensive AI-enabled applications, opting for fully integrated turnkey solutions that help them become more productive faster makes sense. Supermicro is at the vanguard of this movement.

The Supermicro team is ready and well-equipped to design, assemble, test, configure and deploy rack scale solutions. These solutions are ideal for modern datacenter workloads, including AI, deep learning, big data and vSAN.

Why rack scale?

Rack scale solutions let your customers bypass the design, construction and testing of individual servers. Instead of spending precious time and money building, integrating and troubleshooting IT infrastructure, rack scale and cluster-level solutions arrive preconfigured and ready to run.

Supermicro advertises plug-and-play designs. That means your customers need only plug in and connect to their networks, power and optional liquid cooling. After that, it’s all about getting more productivity faster.

Deploying rack scale solutions could enable your customers to reduce or redeploy IT staff, help them optimize their multicloud deployments, and lower their environmental impact and operating costs.

Supermicro + AMD processors = lower costs

Every organization wants to save time and money. Your customers may also need to adhere to stringent environmental, social and governance (ESG) policies to reduce power consumption and battle climate change.

Opting for AMD silicon helps increase efficiency and lower costs. Supermicro’s rack scale solutions feature 4th generation AMD EPYC server processors. These CPUs are designed to shrink rack space and reduce power consumption in your customers’ data center.

AMD says its EPYC-series processors can:

  • Run resource-intensive workloads with fewer servers
  • Reduce operational and energy costs
  • Free up precious data center space and power, then re-allocate this capacity for new workloads and services

Combined with a liquid-cooling system, Supermicro’s AMD-powered rack scale solutions can help reduce your customer’s IT operating expenses by more than 40%.

More than just the hardware

The right rack scale solution is about more than just hardware. Your customers also need a well-designed, fully integrated solution that has been tested and certified before it leaves the factory.

Supermicro provides value-added services beyond individual components to create a rack scale solution greater than the sum of its parts.

You and your customers can collaborate with Supermicro product managers to determine the best platform and components. That includes selecting optimum power supplies and assessing network topology architecture and switches.

From there, Supermicro will optimize server, storage and switch placement at rack scale. Experienced hardware and software engineers will design, build and test the system. They’ll also install mission-critical software benchmarked to your customer’s requirements.

Finally, Supermicro performs strenuous burn-in tests and delivers thoroughly tested L12 clusters to your customer’s chosen site. It’s a one-stop shop that empowers your customers to maximize productivity from day one.

Do more:

 

Featured videos

Follow

Related Content

Supermicro, Vast collaborate to deliver turnkey AI storage at rack scale

Featured content

Supermicro, Vast collaborate to deliver turnkey AI storage at rack scale

Supermicro and Vast Data are jointly offering an AMD-based turnkey solution that promises to simplify and accelerate AI and data pipelines.

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

Supermicro and Vast Data are collaborating to deliver a turnkey, full-stack solution for creating and expanding AI deployments.

This joint solution is aimed at hyperscalers, cloud service providers (CSPs) and large, data-centric enterprises in fintech, adtech, media and entertainment, chip design and high-performance computing (HPC).

Applications that can benefit from the new joint offering include enterprise NAS and object storage; high-performance data ingestion; supercomputer data access; scalable data analysis; and scalable data processing.

Vast, founded in 2016, offers a software data platform that enterprises and CSPs use for data-intensive computing. The platform is based on a distributed systems architecture, called DASE, that allows a system to run read and write operations at any scale. Vast’s customers include Pixar, Verizon and Zoom.

By collaborating with Supermicro, Vast hopes to extend its market. Currently, Vast sells to infrastructure providers at a variety of scales. Some of its largest customers have built 400 petabyte storage systems, and a few are even discussing systems that would store up to 2 exabytes, according to John Mao, Vast’s VP of technology alliances.

Supermicro and Vast have engaged with many of the same CSPs separately, supporting various parts of the solution. By formalizing this collaboration, they hope to extend their reach to new customers while increasing their sell-through to current customers.

Vast is also looking to the Supermicro alliance to expand its global reach. While most of Vast’s customers today are U.S.-based, Supermicro operates in over 100 countries worldwide. Supermicro also has the infrastructure to integrate, test and ship 5,000 fully populated racks per month from its manufacturing plants in California, Netherlands, Malaysia and Taiwan.

There’s also a big difference in size. Where privately held Vast has about 800 employees, publicly traded Supermicro has more than 5,100.

Rack solution

Now Vast and Supermicro have developed a new converged system using Supermicro’s Hyper A+ servers with AMD EPYC 9004 processors. The solution combines 2 separate Vast servers. 

This converged system is well suited to large service providers, where the typical Supermicro-powered Vast rack configuration will start at about 2PB, Mao adds.

Rack-scale configurations can cut costs by eliminating the need for single-box redundancy. This converged design makes the system more scalable and more cost-efficient.

Under the hood

One highlight of the joint project: It puts Vast’s DASE architecture on Supermicro’s  industry-standard servers. Each server will have both the compute and storage functions of a Vast cluster.

At the same time, the architecture is disaggregated via a high-speed Ethernet NVMe fabric. This allows each node to access all drives in the cluster.

The Vast platform architecture uses a series of what the company calls an EBox. Each EBox, in turn, contains 2 kinds of storage servers in a container environment: CNode (short for Compute Node) and DNode (short for Data Node). In a typical EBox, one CNode interfaces with client applications and writes directly to two DNode containers.

In this configuration, Supermicro’s storage servers can act as a hardware building block to scale Vast to hundreds of petabytes. It supports Vast’s requirement for multiple tiers of solid-state storage media, an approach that’s unique in the industry.

CPU to GPU

At the NAB Show, held recently in Las Vegas, Supermicro’s demos included storage servers, each powered by a single-socket AMD EPYC 9004 Series processor.

With up to 128 PCIe Gen 5 lanes, the AMD processor empowers the server to connect more SSDs via NVMe with a single CPU. The Supermicro storage server also lets users move data directly from storage to GPU memory supporting Nvidia’s GPU Direct storage protocol, essentially bypassing a GPU cluster’s CPU using RDMA.

If you or your customers are interested in the new Vast solution, get in touch with your local Supermicro sales rep or channel partner. Under the terms of the new partnership, Supermicro is acting as a Vast integrator and OEM. It’s also Vast’s only rack-scale partner.

Do more:

 

Featured videos

Follow

Related Content

Which media server should you use when you absolutely can’t lose data?

Featured content

Which media server should you use when you absolutely can’t lose data?

A new Linus Tech Tip video shows a real-world implementation of Supermicro storage servers powered by AMD EPYC processors to provide super-high reliability.

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

Are your customers looking for a top-performing media server? And are you looking for a surprisingly entertaining video review of the best one? Then look no further. You’ll find both in the latest Linus Tech Tip video.

This episode, sponsored by Supermicro, is entitled “This Server CANNOT Lose Data.” That gives you an idea of its primary focus: high reliability.

And that reliability is delivered courtesy of a sophisticated server/storage cluster featuring Supermicro GrandTwin A+ multinode servers.

Myriad redundancies

What makes the GrandTwin so reliable? Redundancy. As video host Linus Sebastian exclaims, “Inside this 2U are 4 independent computers!”

Each computer, or node, is powered by a 2.45GHz AMD EPYC processor with up to 128 cores and a 256MB L3 cache. Each node also has 4 front hot-swap 2.5-inch drive bays that can hold petabytes of either NVMe or SATA storage.

The GrandTwin’s nodes can handle up to 3TB of DDR5 ECC server memory. They also have dual M.2 slots for boot drives and 6 PCIe Gen 5 x16 slots for networking, graphics and other expansion cards.

GrandTwin’s high-availability design extends all the way down to its dual power supplies. To ensure the system always has a reliable flow of power to all its vital components, Supermicro added two redundant 2200-watt titanium-level PSUs.

Handling the heat generated by this monster machine is paramount. The GrandTwin takes care of all that hot air via 4 high-speed fans—one fan in each PSU, plus 2 dedicated heavy-duty 8-cm. fans spinning at more than 17,000 RPM.

Prime processing

At the core of each of the GrandTwin’s 4 nodes is an AMD 9004-series processor. Linus’ prized media server, known as “Whonnock 10,” sports an AMD EPYC 9534 CPU in each node.

The EPYC 9534’s cores—there are 64 of them—operate at 2.45GHz and can boost up to 3.7GHz. And because each EPYC processor boasts 12 memory channels, the GrandTwin can address up to 12TB of memory systemwide.

Don’t call it overkill

As Linus says with unbridled enthusiasm, when it comes to redundancy, the name of the game is avoiding “split brain.”

The dreaded split brain can occur when redundant servers have their own object storage. The failure of even a single system can lead to a situation in which each server believes it has the correct data.

If there are only 2 servers, proving which system is correct is impossible. On the other hand, operating 3 or more servers allows the system to resolve the argument and determine the correct data.

Linus and company installed 2 GrandTwin A+ servers. That gives them the 8 redundant systems recommended by their preferred NVMe file system, WEKA.

A multitude of use cases

Your customers may have to contend with thousands of hours of high-resolution videos, like Linus and his cohorts. Or they may develop AI-enabled applications, provide cloud gaming, or host mission-critical web applications.

Whatever the use case, they can benefit from high-reliability servers designed with built-in redundancies. When failure is not an option, your customers need a server that, as the video puts it, “CANNOT lose data.”

That means helping your customers deploy Supermicro GrandTwin A+ servers powered by AMD EPYC processors. It’s the ultimate high-reliability system.

After all, as Linus says, “You only server once.”

Do more:

 

 

Featured videos

Follow

Related Content

How CSPs can accelerate the data center

Featured content

How CSPs can accelerate the data center

A new webinar, now available on demand, offers cloud service providers an overview of new IDC research, outlines roadblocks, and offers guidelines for future success.

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

Are you a cloud services provider—or a CSP wannabe—wondering how to expand your data center in ways that will both keep your customers happy and help you turn a profit?

If so, a recent webinar sponsored by Supermicro and AMD can help. Entitled Accelerate Your Cloud: Best Practices for CSPs, it was moderated by Wendell Wenjen, director of storage market development at Supermicro. Best of all, you can now view this webinar on demand.

Here’s a taste of what you’ll see:

IDC research on CSP buying plans

The webinar’s first speaker is Ashish Nadkarni, group VP and GM of worldwide infrastructure research at IDC. He summarizes new IDC research on technology adoption trends and strategies among service providers.

Sales growth, IDC says, is coming mainly in 4 areas: Infrastructure as a Service (IaaS), hardware (both servers and storage), software and IT services. The good news, Nadkarni adds, is that all 4 can be offered by service providers.

Data centers remain important, Nadkarni says. Not everyone wants to use the public cloud, and not every workload belongs there.

IDC expects that 5 key technologies will be immune to budget cuts:

  • AI and automation
  • Security, risk and compliance
  • Optimization of IT infrastructure and IT operations
  • Back-office applications (HR, SCM and ERP)
  • Customer experience initiatives (for example, chatbots)

Generative AI dominates the conversation, Nadkarni said, and for good reason: IDC expects that this year, GenAI will double the productive use of unstructured data, helping workers discover new insights and knowledge.

Supply-chain issues remain a daunting challenge, IDC finds. Delays can hurt a CSP’s ability to deliver projects, increase the cost of delivering services, and even impair service quality. Owning the supply chain will remain vital.

Other tactics for change, Nadkarni said, include offering a transformation road map; working with a full-stack portfolio provider; and developing a long-term vision for why customers will want to do business with you.

10 steps to data-center scaling

Next up in the webinar is Sim Upadhyayula, VP of solutions enablement at Supermicro. He offered a list of 10 essential steps for scaling a CSP data center.

Topping his list: standardize and scale. There’s no way you can know exactly which workloads will dominate in the future. So be modular. That way, you can scale in smaller increments, keeping customers happy while controlling your costs.

Next on the list: optimize for applications. Unlike big enterprises, most CSPs cannot afford to build application silos. Instead, leading providers will develop an architecture that can cater to all. That means using standard hardware that can later be optimized for specific workloads.

Common challenges

Suresh Andani, AMD’s senior director of product management for server cloud, is up next. He discusses 3 key CSP challenges:

  • Market disruption: Caused by a changing ISV landscape, and by increasing power and cooling costs.
  • Aging infrastructure: Service providers with older systems find them costly to maintain, unable to keep pace with customers’ business demands, and vulnerable to increasingly dangerous security threats.
  • Expanding demands: Customers keep raising the bar on core workloads AI, cloud-native applications, digital transformation, the hybrid workforce and security enhancements.

During the webinar’s concluding roundtable discussion, Andani also emphasized the importance of marrying the right infrastructure with your workloads. That way, he said, CSPs can operate efficiently, making the most of their power and compute cycles.

“Work with your vendors to provide the best compute solutions,” Andani of AMD advised. “Later you can offer a targeted infrastructure for high performance compute, another for enterprise workloads, another for gaming, and another for rendering.”

Lean on your providers, he added, to provide the right solution, whether your target is performance or cost.

Do more:

 

Featured videos

Follow

Related Content

Tech Explainer: What’s the difference between AI training and AI inference?

Featured content

Tech Explainer: What’s the difference between AI training and AI inference?

AI training and inference may be two sides of the same coin, but their compute needs can be quite different. 

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

Artificial Intelligence (AI) training and inference are two sides of the same coin. Training is the process of teaching an AI model how to perform a given task. Inference is the AI model in action, drawing its own conclusions without human intervention.

Take a theoretical machine learning (ML) model designed to detect counterfeit one-dollar bills. During the training process, AI engineers would feed the model large data sets containing thousands, or even millions, of pictures. And tell the training application which are real and which are counterfeit.

Then inference could kick in. The AI model could be uploaded to retail locations, then run to detect bogus bills.

A deeper look at training

That’s the high level. Let’s dig in a bit deeper.

Continuing with our bogus-bill detecting workload, during training, the pictures fed to the AI model would include annotations telling the AI how to think about each piece of data.

For instance, the AI might see a picture of a dollar bill with an embedded annotation that essentially tells the model “this is legal tender.” The annotation could also identify characteristics of a genuine dollar, such as the minute details of the printed iconography and the correct number of characters in the bill’s serial number.

Engineers might also feed the AI model pictures of counterfeit bills. That way, the model could learn the tell-tale signs of a fake. These might include examples of incomplete printing, color discrepancies and missing watermarks.

On to inference

One the training is complete, inference can take over.

Still with our example of counterfeit detection, the AI model could now be uploaded to the cloud, then connected with thousands of point-of-sale (POS) devices in retail locations worldwide.

Retail workers would scan any bill they suspect might be fake. The machine learning model, in turn, would then assess the bill’s legitimacy.

This process of AI inference is autonomous. In other words, once the AI enters inference, it’s no longer getting help from engineers and app developers.

Using our example, during inference the AI system has reached the point where it can reliably discern both legal and counterfeit bills. And it can do so with a high enough success percentage to satisfy its human controllers.

Different needs

AI training and inference also have different technology requirements. Basically, training is far more resource-intensive. The focus is on achieving low-latency operation and brute force.

Training a large language model (LLM) chatbot like the popular ChatGPT often forces its underlying technology to contend with more than a trillion parameters. An AI parameter is a variable learned by the LLM during training. These parameters include configuration settings and components that define the LLM’s behavior.)

To meet these requirements, IT operations must deploy a system that can bring to bear raw computational power in a vast cluster.

By contrast, inference applications have different compute requirements. “Essentially, it’s, ‘I’ve trained my model, now I want to organize it,’” explained AMD executive VP and CTO Mark Papermaster in a recent virtual presentation.

AMD’s dual-processor solution

Inferencing workloads are both more concise and less demanding than those for training. Therefore, it makes sense to run them on more affordable GPU-CPU combination technology like the AMD Instinct MI300A.

The AMD Instinct MI300A is an accelerated processing unit (APU) that combines the facility of a standard AI accelerator with the efficiency of AMD EPYC processors. Both the CPU and GPU elements can share memory, dramatically enhancing efficiency, flexibility and programmability.

A single AMD MI300A APU packs 228 GPU compute units, 24 of AMD’s ‘Zen 4’ CPU cores, and 128GB of unified HBM3 memory. Compared with the previous-generation AMD MI250X accelerators, this translates to approximately 2.6x the workload performance per watt using FP32.

That’s a significant increase in performance. It’s likely to be repeated as AI infrastructure evolves along with the proliferation of AI applications that now power our world.

Do more:

 

 

Featured videos

Follow

Related Content

Research Roundup: Tech leaders’ time, GenAI for HR, network security in the cloud, dangerous dating sites

Featured content

Research Roundup: Tech leaders’ time, GenAI for HR, network security in the cloud, dangerous dating sites

Catch up on the latest IT market research from MIT, Gartner, Dell’Oro Group and others. 

Learn More about this topic
  • Applications:

Tech leaders are spending more time with the channel. HR execs are getting serious about GenAI. Network security is moving to the cloud. And online dating sites can be dangerous.

That’s some of the latest and greatest from leading IT researchers. And here’s your Performance Intensive Computing roundup.

Tech leaders spend more time with the channel

C-level technology executives in 2022 spent 17% of their time working with external customers and channel partners, up from 10% of their time in 2007, according to a recent report from the MIT Center for Information Systems Research (CISR).

Good news, right? Well, conversely, these same tech leaders spent less time collaborating with their coworkers and working on their organizations’ technology stacks. Guess something had to give.

The report, published earlier this year, is the work of three MIT researchers. To compile the data, they reviewed surveys of CIOs, CTOs and CDOs conducted in 2007, 2016 and 2022.

Why the 7% increase in time spent with external customers and channel partners? According to the MIT researchers, it’s the “growing number of digital touchpoints.”

HR execs using GenAI

Nearly 4 in 10 HR executives (38%) are now piloting, planning to implement, or have already implemented generative AI, finds research firm Gartner. That’s up sharply from just 19% of HR execs as recently as last June.

The results come from a quick Gartner survey. This past Jan. 31, the firm polled nearly 180 HR execs.

One of the survey’s key findings: “More organizations are moving from exploring how GenAI might be used…to implementing solutions,” says Dion Love, a VP in Gartner’s HR practice.

Gartner’s January survey also found 3 top use cases for GenAI in HR:

  • HR service delivery: Of those working with GenAI, over 4 in 10 (43%) are using the technology for employee-facing chatbots.
  • HR operations: Nearly as many (42%) are working with GenAI for administrative tasks, policies and generating documents.
  • Recruiting: About the same percentage (41%) are working with GenAI for job descriptions and skills data.

Yet all this work is not leading to many new GenAI-related job roles. Over two-thirds of the respondents (67%) said they do not plan to add any GenAI-related roles to the HR function over the next 12 months.

Network security moving to the cloud

Sales of SaaS-based and virtual network-security solutions surged last year by 26%, reaching a global total of $9.6 billion. By contrast, the overall network-security market shrank by 1%.

That’s according to a report from Dell’Oro Group. It calls the move to network-security solutions in the cloud a “pivotal shift.”

Dell’Oro senior director Mauricio Sanchez goes even further. He calls the industry’s gravitation toward SaaS and virtual solutions “nothing short of revolutionary.”

Also, nearly $5 billion of that $9.6 billion market was due to a 30% rise in spending on SSE networks, Dell’Oro says. SSE, short for Security Service Edge, incorporates various services—including network service brokering, identity service brokering, and security as a service—in a single package.

Looking for love online? Be careful

Nearly 7 out of 10 online daters have been scammed while using dating sites. Some of the victims lost money; others risked their personal security.

That’s according to a survey conducted for ID Shield. The survey was limited, reaching only about 270 people. But all respondents had at least used a dating app in the last 3 years.

The survey’s key findings:

  • Financial loss: Six in 10 scam victims on dating sites lost more than $10,000 to the crooks. And slightly more (64%) disclosed personal and finance information that was later used against them.
  • ID theft: Nearly 7 in 10 respondents were asked to verify their identity to someone on the dating app. And nearly two-thirds (65%) divulged their Social Security numbers. 
  • Repeat users: You might think the victims would learn. But 93% of users who were scammed once on a dating app say they continue to use the same app. Let’s hope they’re at least being careful.

 

Featured videos

Follow

Related Content

10 best practices for scaling the CSP data center — Part 1

Featured content

10 best practices for scaling the CSP data center — Part 1

Cloud service providers, here are best practices—courtesy of Supermicro—to help you design and deploy rack-scale data centers. 

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

Cloud service providers, here are 10 best practices—courtesy of Supermicro—that you can follow for designing and deploying rack-scale data centers. All are based on Supermicro’s real-world experience with customers around the world.

Best Practice No. 1: First standardize, then scale

First, select a configuration of compute, storage and networking. Then scale these configurations up and down into setups you designate as small, medium and large.

Later, you can deploy these standard configurations at various data centers with different numbers of users, workload sizes and growth estimates.

Best Practice No. 2: Optimize the configuration

Good as Best Practice No. 1 is, it may not work if you handle a very wide range of workloads. If that’s the case, then you may want to instead optimize the configuration.

Here’s how. First, run the software on the rack configuration to determine the best mix of CPUs, including cores, memory, storage and I/O. Then consider setting up different sets of optimized configurations.

For example, you might send AI training workloads to GPU-optimized servers. But a database application on a standard 2-socket CPU system.

Best Practice No. 3: Plan for tech refreshes 

When it comes to technology, the only constant is change itself. That doesn’t mean you can just wait around for the latest, greatest upgrade. Instead, do some strategic planning.

That might mean talking with key suppliers about their road maps. What are their plans for transitions, costs, supply chains and more?

Also consider that leading suppliers now let you upgrade some server components without having to replace the entire chassis. That reduces waste. That could also help you get more power from your current racks, servers and power requirements.

Best Practice No. 4: Look for new architectures

New architectures can help you increase power at lower cost. For example, AMD and Supermicro offer data-center accelerators that let you run AI workloads on a mix of GPUs and CPUs, a less costly alternative to all-GPU setups.

To find out if you could benefit from new architectures, talk with your suppliers about running proof-of-concept (PoC) trials of their new technologies. In other words, try before you buy.

Best Practice No. 5: Create a support plan

Sure, you need to run 24x7, but that doesn’t mean you have to pay third parties for all of that. Instead, determine what level of support you can provide in-house. For what remains, you can either staff up or outsource.

When you do outsource, make sure your supplier has tested your software stack before. You want to be sure that, should you have a problem, the supplier will be able to respond quickly and correctly.

Do more:

 

Featured videos

Follow

Related Content

10 best practices for scaling the CSP data center — Part 2

Featured content

10 best practices for scaling the CSP data center — Part 2

Cloud service providers, here are more best practices—courtesy of Supermicro—that you can follow for designing and deploying rack-scale data centers. 

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

Cloud service providers, here are 5 more best practices—courtesy of Supermicro—that you can follow for designing and deploying rack-scale data centers. All are based on Supermicro’s real-world experience with customers around the world.

Best Practice No. 6: Design at the data-center level

Consider your entire data center as a single unit, complete with its range of both strengths and weaknesses. This will help you tackle such macro-level issues as the separation of hot and cold aisles, forced air cooling, and the size of chillers and fans.

If you’re planning an entirely new data center, remember to include a discussion of cooling tech. Why? Because the physical infrastructure needed for an air-cooled center is quite different than that needed for liquid cooling.

Best Practice No. 7: Understand & consider liquid cooling

We’re approaching the limits of air cooling. A new approach, one based on liquid cooling, promises to keep processors and accelerators running within their design limits.

Liquid cooling can also reduce a data center’s Power Usage Effectiveness (PUE) ratio, a measure of how much energy is used by a center’s computing equipment. This cooling tech can also minimize the need for HVAC cooling power.

Best Practice No. 8: Measure what matters

You can’t improve what you don’t measure. So make sure you are measuring such important factors as your data center’s CPU, storage and network utilization.

Good tools are available that can take these measurements at the cluster level. These tools can also identify both bottlenecks and levels of component over- or under-use.

Best Practice No. 9: Manage jobs better

A CSP’s data center is typically used simultaneously by many customers. That pretty much means using a job-management scheduler tool.

One tricky issue is over-demand. That is, what do you do if you lack enough resources to satisfy all requests for compute, storage or networking? A job scheduler can help here, too.

Best Practice No. 10: Simplify your supply chain

Sure, competition across the industry is a good thing, driving higher innovation and lower prices. But within a single data center, standardizing on just a single supplier could be the winning ticket.

This approach simplifies ordering, installation and support. And if something should go wrong, then you’ll have only the proverbial “one throat to choke.”

Can you still use third-party hardware as appropriate? Sure. And with a single main supplier, that integration should be simpler, too.

Do more:

 

Featured videos

Follow

Related Content

Data-center service providers: ready for transformation?

Featured content

Data-center service providers: ready for transformation?

An IDC researcher argues that providers of data-center hosting services face new customer demands that require them to create new infrastructure stacks. Key elements will include rack-scale integration, accelerators and new CPU cores. 

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

If your organization provides data-center hosting services, brace yourself. Due to changing customer demands, you’re about to need an entirely new infrastructure stack.

So argues Chris Drake, a senior research director at market watcher IDC, in a recently published white paper sponsored by Supermicro and AMD, The Power of Now: Accelerate the Datacenter.

In his white paper, Drake asserts that this new data center infrastructure stack will include new CPU cores, accelerated computing, rack-scale integration, a software-defined architecture, and the use of a micro-services application environment.

Key drivers

That’s a challenging list. So what’s driving the need for this new infrastructure stack? According to Drake, changing customer requirements.

More specifically, a growing need for hosted IT requirements. For reasons related to cost, security and performance, many IT shops are choosing to retain proprietary workloads on premises and in private-cloud environments.

While some of these IT customers have sufficient capacity in their data centers to host these workloads on prem, many don’t. They’ll rely instead on service providers for a range of hosted IT requirements. To meet this demand, Drake says, service providers will need to modernize.

Another driver: growing customer demand for raw compute power, a direct result of their adoption of new, advanced computing tools. These include analytics, media streaming, and of course the various flavors of artificial intelligence, including machine learning, deep learning and generative AI.

IDC predicts that spending on servers ranging in price from $10K to $250K will rise from a global total of $50.9 billion in 2022 to $97.4 billion in 2027. That would mark a 5-year compound annual growth rate of nearly 14%.

Under the hood

What will building this new infrastructure stack entail? Drake points to 5 key elements:

  • Higher-performing CPU cores: These include chiplet-based CPU architectures that enable the deployment of composable hardware architectures. Along with distributed and composable hardware architectures, these can enable more efficient use of shared resources and more scalable compute performance.
  • Accelerated computing: Core CPU processing will increasingly be supplemented by hardware accelerators, including those for AI. They’ll be needed to support today’s—and tomorrow’s—increasingly diverse range of high-performance and data-intensive workloads.
  • Rack-scale integration: Pre-tested racks can facilitate faster deployment, integration and expansion. They can also enable a converged-infrastructure approach to building and scaling a data center.
  • Software-defined data center technology: In this approach, virtualization concepts such as abstraction and pooling are extended to a data center’s compute, storage, networking and other resources. The benefits include increased efficiency, better management and more flexibility.
  • A microservices application architecture: This approach divides large applications into smaller, independently functional units. In so doing, it enables a highly modular and agile way for applications to be developed, maintained and upgraded.

Plan for change

Rome wasn’t built in a day. Modernizing a data center will take time, too.

To help service providers implement a successful modernization, Drake of IDC offers this 6-point action plan:

1. Develop a transformation road map: Aim to strike a balance between harnessing new technology opportunities on the one hand and being realistic about your time frames, costs and priorities on the other.

2. Work with a full-stack portfolio vendor: You want a solution that’s tailored for your needs, not just an off-the-rack package. “Full stack” here means a complete offering of servers, hardware accelerators, storage and networking equipment—as well as support services for all of the above.

3. Match accelerators to your workloads: You don’t need a Formula 1 race car to take the kids to school. Same with your accelerators. Sure, you may have workloads that require super-low latency and equally high thruput. But you’re also likely to be supporting workloads that can take advantage of more affordable CPU-GPU combos. Work with your vendors to match their hardware with your workloads.

4. Seek suppliers with the right experience: Work with tech vendors that know what you need. Look for those with proven track records of helping service providers to transform and scale their infrastructures.

5. Select providers with supply-chain ownership: Ideally, your tech vendors will fully own their supply chains for boards, systems and rack designs such as liquid-cooling systems. That includes managing the vertical integration needed to combine these elements. The right supplier could help you save costs and get to market faster.

6. Create a long-term plan: Plan for the short term, but also look ahead into the future. Technology isn’t sitting still, and neither should you. Plan for technology refreshes. Ask your vendors for their road maps, and review them. Decide what you can support in-house versus what you’ll probably need to hand off to partners.

Now do more:

 

Featured videos

Follow

Related Content

Pages