Sun Cofounder Evangelizes Liquid Blade Server

By Michael Feldman

May 4, 2011

What does a Sun Microsystems cofounder do with his spare time? Well, if you’re Scott McNealy, you spend some if it lending your expertise to promising tech vendors that are looking to break into the IT big leagues. One such company that he has taken a personal interest in is Hardcore Computer, which recently introduced a line of servers that use liquid submersion technology. HPCwire spoke with McNealy to get his take on the technology and to ask him why he thinks the company deserves the spotlight.

McNealy signed on as a non-paid advisor and consultant with Hardcore in January at the behest of longtime friend and former Stanford classmate Doug Burgum. Burgum’s venture firm, Kilbourne Group, has invested in Hardcore, a Rochester, Minnesota-based computer maker that specializes in high performance gear based on the company’s patented liquid submersion cooling technology.

“This is one of the few companies innovating on top of the Intel architecture — rather than just strapping a power supply on and porting Linux,” McNealy told HPCwire.

Hardcore makes a range of liquid-cooled offerings, including desktops, workstations, and servers. Its latest offering is the “Liquid Blade,” a server line the company announced in May 2010 and launched in November at the Supercomputing Conference in New Orleans (SC10). The new blade is more or less a standard dual-socket x86-based blade using Intel Xeon 5500 and 5600 Xeon parts. It sports eight DDR3 memory slots per CPU, six SATA slots for storage, and a PCIe x16 slot for a GPU card or other external device.

Liquid Blade’s secret sauce — and in this case it literally is a sauce — is Hardcore’s patented liquid submersion technology. The company uses a proprietary dielectric fluid, called Core Coolant, to entirely submerge the blades within a specially-built 5U rack-mounted chassis. The coolant is inert, biodegradable, and most importantly non-conductive, so all of the electrical components inside the server are protected.

As with any liquid coolant, the idea is to draw off the excess heat much more efficiently than an air-cooled setup and ensure all the server components are keep comfortably cool even under maximum load. According to the company literature, the Core Coolant has 1,350 times the cooling capacity of air. Since the coolant is so effective at heat dissipation and the internal fans have been dispensed with, the server components can be packed rather densely. In this case the 5U Hardcore chassis can house up to seven of the dual-socket blades.

The company launched its liquid-dipped server at SC10 last November to get the attention of the HPC community, but the offering is suitable for any installation where the datacenter is constrained by power and space. Besides HPC centers, these include DoD facilities, telco firms, and Internet service providers. “You have to look at the users who think at scale and have a huge electric bill,” explains McNealy.

The datacenter cooling problem is well-known, of course. As servers get packed with hotter and faster chips and datacenters scale up to meet growing demand, getting enough power and space has become increasingly challenging. Datacenter cooling has traditionally relied on air conditioning, but air makes for a poor heat exchange medium, and it’s hard to direct it where it’s most needed. “Air goes everywhere but where you want it to,” laughs McNealy. Cooling a hot server, he says is “like trying to blow a candle out from the other side of the room.”

Because of the density of the Hardcore solution, you need about 50 percent fewer racks to deliver the same compute. And since the blades essentially never overheat, one can expect better reliability and longevity. As any datacenter administrator knows, heat is a major cause of server mortality, especially in facilities filled to capacity.

But the really big savings is on the power side. Since cooling and the associated equipment take up such a large chunk of a datacenter energy budget, any effort to reduce these costs tends to pay for itself in just a few years. An independent study found that a Liquid Blade setup could reduce datacenter cooling costs by up to 80 percent and operating costs by up to 25 percent.

Hardcore isn’t alone in the liquid submersion biz. Other companies, most notably Austin-based Green Revolution, are providing these types of products. In the case of Green Revolution, they offer a general-purpose solution for all sorts of hardware — rack servers, blades, and network switches. The company will strip down the gear to its essentials and immerse the components in a specially-built 42U enclosure filled with an inert mineral oil.

But since Hardcore is dunking its own servers, it has the option to build high performance gear that would be impractical to run in an air-cooled environment. As McNealy points out, the efficient liquid cooling is a natural for the highest bin x86 chips running the fastest clocks. For example, the company could stuff Intel’s latest 4.4 GHz Xeon 5600 processors into its blades, and offer a special-purpose product for high frequency traders (as Appro has done, sans immersive liquid cooling, with its HF1 servers). Hardcore has never talked about such a setup for HFT, but it does tout the servers outfitted with high wattage graphics cards for GPGPU type computation. Applications using such capabilities include medical imaging, CGI rendering, engineering simulation and modeling and web-based gaming.

One the things McNealy has been working with the Hardcore people on is getting an apples-to-apples comparison of their liquid cooled gear versus conventional air-cooled servers. To do this, he says, you have come up with a higher level analysis that takes into account the service cost over the entire datacenter.

According to the company, the cost of a Liquid Blade setup is on par with a comparably equipped air-cooled product since all the fans are eliminated and the chassis design is simpler. If a user opted for Liquid Blade when it came time to upgrade their servers, they could start to realize energy costs savings immediately. But the big savings occur when a datacenter can be built from scratch with liquid submersion in mind.

In that case, the datacenter can dispense with a lot of the CRAC units, use 12-foot ceilings instead of 16-foot ones (no overhead air ductwork is needed), and use less UPS units thanks to reduced power requirements. The only extra cost comes with the chilled water to oil heat exchangers used to draw the heat from the chassis coolant. Also, since you can fit more servers into the same space, the datacenter floor space can be reduced by about 30 percent for a given compute capacity.

So why isn’t everyone flocking to liquid submersion? Customer inertia, says McNealy. According to him, he’s spent most of his career knowing the right thing to do and trying to get others to realize it themselves.

With Hardcore, the challenge is that most organizations are already set up with their existing air-cooled facilities, so a lot of the cost incentives for the big switch aren’t there. He thinks if a large Internet service provider bought into this technology for a new datacenter, the business could quickly take off. For McNealy’s Sun, that tipping point was in the late 80s when Computervision made a big deal to go with his company’s Unix-based workstations. Hardcore, no doubt would love to repeat history, this time with the likes of Google, Amazon, or Facebook.

“Their biggest challenges is the barrier to exit from the old strategy, not the barrier to entry to the new one,” says McNealy.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

2024 Winter Classic: Meet Channel Islands “A”

May 3, 2024

This is the second team from California State University, Channel Islands – or maybe it’s the first team? Not sure, but I do know they have two teams total, and this is one of them. As you’ll see in the video in Read more…

Intersect360 Research Takes a Deep Dive into the HPC-AI Market in New Report

May 3, 2024

A new report out of analyst firm Intersect360 Research is shedding some new light on just how valuable the HPC and AI market is. Taking both of these technologies as a singular unit, Intersect360 Research found that the Read more…

Hyperion To Provide a Peek at Storage, File System Usage with Global Site Survey

May 3, 2024

Curious how the market for distributed file systems, interconnects, and high-end storage is playing out in 2024? Then you might be interested in the market analysis that Hyperion Research is planning on rolling out over Read more…

2024 Winter Classic: Meet Team Jackson State

May 3, 2024

This is the second time we’re seeing a team from Jackson State university. The team features two veterans of the 2023 Winter Classic, which should help, but it’s also a team whose members are involved in a lot of oth Read more…

2024 Winter Classic: NASA Results Revealed!

May 2, 2024

In this edition of the Winter Classic Studio Update Show we reveal the results from the NASA BTIO Challenge. The benchmark, BTIO, is a subset of the NAS Parallel benchmark and NASA set up a formidable set of milestones, Read more…

2024 Winter Classic: NASA Mentor Interview

May 2, 2024

The folks at NASA Ames once again did a bang-up job as a mentor for the 2024 Winter Classic. This is the third time they’ve fulfilled this vital function, and their challenges keep getting better and better. In thei Read more…

Hyperion To Provide a Peek at Storage, File System Usage with Global Site Survey

May 3, 2024

Curious how the market for distributed file systems, interconnects, and high-end storage is playing out in 2024? Then you might be interested in the market anal Read more…

Qubit Watch: Intel Process, IBM’s Heron, APS March Meeting, PsiQuantum Platform, QED-C on Logistics, FS Comparison

May 1, 2024

Intel has long argued that leveraging its semiconductor manufacturing prowess and use of quantum dot qubits will help Intel emerge as a leader in the race to de Read more…

Stanford HAI AI Index Report: Science and Medicine

April 29, 2024

While AI tools are incredibly useful in a variety of industries, they truly shine when applied to solving problems in scientific and medical discovery. Research Read more…

IBM Delivers Qiskit 1.0 and Best Practices for Transitioning to It

April 29, 2024

After spending much of its December Quantum Summit discussing forthcoming quantum software development kit Qiskit 1.0 — the first full version — IBM quietly Read more…

Shutterstock 1748437547

Edge-to-Cloud: Exploring an HPC Expedition in Self-Driving Learning

April 25, 2024

The journey begins as Kate Keahey's wandering path unfolds, leading to improbable events. Keahey, Senior Scientist at Argonne National Laboratory and the Uni Read more…

Quantum Internet: Tsinghua Researchers’ New Memory Framework could be Game-Changer

April 25, 2024

Researchers from the Center for Quantum Information (CQI), Tsinghua University, Beijing, have reported successful development and testing of a new programmable Read more…

Intel’s Silicon Brain System a Blueprint for Future AI Computing Architectures

April 24, 2024

Intel is releasing a whole arsenal of AI chips and systems hoping something will stick in the market. Its latest entry is a neuromorphic system called Hala Poin Read more…

Anders Dam Jensen on HPC Sovereignty, Sustainability, and JU Progress

April 23, 2024

The recent 2024 EuroHPC Summit meeting took place in Antwerp, with attendance substantially up since 2023 to 750 participants. HPCwire asked Intersect360 Resear Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Leading Solution Providers

Contributors

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

Eyes on the Quantum Prize – D-Wave Says its Time is Now

January 30, 2024

Early quantum computing pioneer D-Wave again asserted – that at least for D-Wave – the commercial quantum era has begun. Speaking at its first in-person Ana Read more…

The GenAI Datacenter Squeeze Is Here

February 1, 2024

The immediate effect of the GenAI GPU Squeeze was to reduce availability, either direct purchase or cloud access, increase cost, and push demand through the roof. A secondary issue has been developing over the last several years. Even though your organization secured several racks... Read more…

Intel Plans Falcon Shores 2 GPU Supercomputing Chip for 2026  

August 8, 2023

Intel is planning to onboard a new version of the Falcon Shores chip in 2026, which is code-named Falcon Shores 2. The new product was announced by CEO Pat Gel Read more…

GenAI Having Major Impact on Data Culture, Survey Says

February 21, 2024

While 2023 was the year of GenAI, the adoption rates for GenAI did not match expectations. Most organizations are continuing to invest in GenAI but are yet to Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire