HPE Armed And Ready For The Enterprise And Service Providers With Ampere-Based HPE ProLiant RL300 Gen11 Server 

By Patrick Moorhead - July 25, 2022

HPE Discover 2022 started with a bang today as HPE announced the ProLiant RL300 Gen 11 server. Moor Insights & Strategy’s Patrick Moorhead and I were pre-briefed on this announcement by Executive Vice President and GM of HPE's Compute Business, Neil MacDonald. In the following paragraphs, Pat and I will cover what was announced and our analysis of this move. It’s important to know that this is the initial announcement, not the launch, so every detail isn’t available now.

The news

The HPE ProLiant RL300 Gen 11 server is based on Ampere’s silicon, designed for service providers and enterprise organizations deploying cloud-native environments. Ampere is the leading developer of merchant, Arm-based server CPUs with customers such as Oracle Cloud, Azure, and Chinese cloud giant Tencent. AWS is on its third-generation Arm-based Graviton3.

Customers can purchase this single-socket server with either the Ampere Altra or Altra Max CPU, supporting up to 128 single-threaded cores and up to 16 DIMM slots for up to 4TB of RAM. Additionally, the HPE ProLiant RL300 Gen11 server will support 3 PCIe Gen4 expansion slots and 2 Open Compute Project (OCP) 3.0 capable slots. From a management and security perspective, the server will ship with HPE iLO, HPE’s silicon root of trust technology, and with the ability to enable OpenBMC, offering customers the flexibility to pursue their open-source strategy. This openness is critical for servicing the service provider space.

HPE ProLiant RL300 HPE

When looking at the details around the HPE ProLiant RL300 Gen11 server, it is clear HPE has targeted scale-out, cloud-native environments. Its single socket design with a richness of memory, storage, one thread-one instance, and support for accelerators such as GPUs indicates a server that should be able to handle many of the workloads that power the digitally transformed enterprise. Or the service provider (SP) that is supporting these cloud-native environments.

It is also clear that the RL300 is a first-class citizen in the HPE ProLiant portfolio, with all of the management and security capabilities that are hallmarks of HPE-designed servers.

Customers can acquire the HPE ProLiant RL300 Gen11 server directly, through HPE, and the channel. The new server will also be available as-a-service through HPE GreenLake. While details around specific HPE GreenLake supported services weren’t discussed, I suspect there will be a lot of customer demand to deliver RL300-based cloud-native solution stacks that can be quickly deployed and provisioned.

Why build an Arm-based server? Defining the problem statement

Before getting into more detail around the RL300, it’s worth a paragraph or two talking about why HPE would invest in an Arm-based server. There are a couple of market dynamics to unpack when considering this move. The first is the explosion of Arm in the cloud service provider (CSP) space. The likes of AWS, Azure, and Oracle Cloud deploying Arm-based instances is not just a trend – it’s likely an inflection point. These environments are growing like wildfire as cloud consumers (enterprise IT) find cost savings and maybe even energy savings without a corresponding performance hit.

On the other side of this equation, the CSPs find Arm attractive due to cost savings – both direct (acquisition) and, more importantly, operationally (primarily driven by power savings). Any IT person who has managed a large-scale datacenter understands what this means. When looking at real TCO numbers, the costs associated with powering a datacenter full of servers dwarfs the cost of acquiring that hardware.

Modern technology inflection points start in the cloud, gain scale, and then move into the enterprise datacenter. And this Arm revolution is no exception. Digitally transformed enterprises want to achieve the same economic benefits as the CSPs they utilize. And for those cloud-native projects that are rapidly growing, cloud-native infrastructure only makes sense.

Addressing the growing challenges in the modern datacenter with the ProLiant RL300 HPE

HPE, embracing its first-mover spirit, has recognized this inflection point and is delivering on the front side of the demand curve. Let’s be clear – this is a significant investment. Top-tier OEMs invest millions of dollars in non-recurring engineering (NRE) budgets and countless resources in bringing a new server platform to market. Because of this high up-front cost, such decisions are not made lightly. I say this to underscore my position that this announcement is more significant than some may understand. HPE is making a big statement by leading its Gen11 launch wave with a server powered by Ampere’s Arm-based Altra CPU.

As shown in the above graphic, there are two primary customer types for the ProLiant RL300- enterprise IT organizations looking to replicate the cloud on-premises, and the service provider community looking to deliver cloud-native services to a customer base in more sustainable and economical ways.

These enterprises and service provider firms with CSP-like operational aspirations are unwilling to take CSP-like risks with infrastructure. What does this mean? CSPs invest in white box server technology from original design manufacturers (ODMs) that have quality standards far below what an enterprise or service provider tolerable. They do this because it leads to savings over the lifecycle of that deployed infrastructure. Service providers and enterprise IT organizations don’t have this same risk tolerance, so a company like HPE enables that cloud operating model minus the risk.

My take on the HPE ProLiant RL300 Gen11 server

I am bullish on Arm in enterprise cloud environments. Mainly for the reasons I articulated previously – enterprise adopts cloud technologies in a trailing fashion. And the strong uptake of Arm across the major cloud players has left a lot of IT strategists asking, “why not?”

I am also bullish on HPE as an IT solutions provider that will power the next-generation datacenter. It is a company that seems to have a knack for knowing what's next and getting out in front of those trends. It doesn’t get it always right, but I believe the past few years it has made the right moves. Just travel back to Discover of 2019 when Antonio Neri stood up and announced HPE GreenLake and the intent to deliver the entire HPE portfolio as-a-Service. It was the beginning of this consumption-based IT market as we know it and now everybody is doing it.

I believe the RL300 strikes a good balance between meeting the needs of the digitally transformed enterprise looking to embrace openness in a fully managed way. I like that HPE fully considers the diversity of the customers that will deploy RL300. For example, the RL300 will ship with both iLO and OPENBMC for management capabilities. This allows traditional HPE customers to manage these servers with the HPE toolchain and accounts for potential customers deploying the RL300 in a more open environment. It is these seemingly less visible enablements that demonstrate HPE’s smartness in design.

What about the x86 players?

Reading this analysis, one could walk away thinking we see the world moving to Arm. To be clear, this is not the view of myself, Moor Insights & Strategy, HPE, Ampere, or any other entity. Each architecture has its place in the datacenter. Arm has found some use cases that shine quite well (web serving, caching, media transcoding, some database, etc.). And these use cases served by Arm are well supported through a rich ecosystem of software players. The appeal of Arm in these workloads is the notion of single threaded performance that enables predictable, dedicated performance.

But this doesn't mean that organizations will replace x86 with Arm. Instead, we see this as Arm complementing x86 in supporting the digitally transformed enterprise. And while I don't work for HPE's marketing or positioning groups, I am sure these teams will be very prescriptive in where they position the ProLiant RL300 and how.

AMD and Intel will undoubtedly continue to target Arm and companies like Ampere as they become more significant blips on the strategic radar screen. And this competition and resulting innovation are good for all. Especially the customer.

Final Thoughts

The "Arm in the datacenter" debate has gone on for years. Twelve years by my count as we pontificated on Calxeda’s 32-bit solutions. During those years, we have all witnessed a lot of startups (and established players) in the semiconductor space begin and fail at delivering a viable Arm-based CPU. Part of these failings had to do with architectures that were not fully enterprise-ready or not performant enough. And part of these failings had to do with a software ecosystem that wasn't entirely built and mature enough. And finally, part of this was a hardware ecosystem that had not yet seen the market/customer need.

These factors no longer exist. Our founder and CEO, Patrick Moorhead, cites AWS’s Graviton3 as the real game changer in the space. Ampere brought a team of industry players who understood what was necessary to deliver an enterprise-class CPU based to market. And, I believe, they delivered.

The software ecosystem is vibrant. Arm is a first-class citizen in the Linux community and the Cloud Native Computing Foundation (CNCF). 

And HPE sees the market need and opportunity and is addressing these with the announcement of the ProLiant RL300 Gen11 server. This server is slated to ship in calendar Q3 of 2022. You can be sure we'll be checking back to see how HPE handles its go-to-market (GTM) efforts, along with detailed performance benchmarking that compares and contrasts with other architectures. 

Stay tuned.

Note: This analysis was co-written by Patrick Moorhead and Matt Kimball

+ posts
Patrick founded the firm based on his real-world world technology experiences with the understanding of what he wasn’t getting from analysts and consultants. Ten years later, Patrick is ranked #1 among technology industry analysts in terms of “power” (ARInsights)  in “press citations” (Apollo Research). Moorhead is a contributor at Forbes and frequently appears on CNBC. He is a broad-based analyst covering a wide variety of topics including the cloud, enterprise SaaS, collaboration, client computing, and semiconductors. He has 30 years of experience including 15 years of executive experience at high tech companies (NCR, AT&T, Compaq, now HP, and AMD) leading strategy, product management, product marketing, and corporate marketing, including three industry board appointments.