Coming Soon: Virtualized Everything

By By Derrick Harris, Editor

May 5, 2008

With IT expenditures both capital (CAPEX) and operational (OPEX) spiraling out of control, businesses have swarmed to virtualization to help consolidate servers, cut energy costs and maximize utilization, among other benefits. But the server virtualization being offered by vendors like VMware is just the tip of the clichéd iceberg.

With their collective eyes set on drastically cutting CAPEX and OPEX while significantly increasing flexibility and scalability, a new breed of vendors is targeting the rest of the datacenter for virtualization. One of these companies is 3Leaf Systems, a Silicon Valley startup with a simple vision: to take a pool of x86 servers and decompose them into I/O, compute and memory, and make each available on demand. 3Leaf has a two-phase Virtual Compute Environment approach, says Senior Director of Marketing Rob Reiner, starting with I/O virtualization (currently available) and culminating with compute and memory virtualization spanning across physical machines (available beginning in 2009).

It Begins with I/O

The company’s virtual I/O software strips local storage, NICs and HBAs and replaces them with virtual versions. The servers are then connected via InfiniBand (although Ethernet functionality is on the way) to a commodity switch fabric, which is connected to a pair of 3Leaf’s V8000 virtual I/O servers. The V8000s connect directly to the SAN and LAN, and the entire setup requires approximately 80 percent fewer standby servers and 80 percent fewer SAN and LAN ports. According to Reiner, it also has saved proof-of-concept customers an average of 50 percent in CAPEX and OPEX.

Speaking specifically about OPEX, Reiner says the big savings come because server configurations are saved as profiles, so provisioning is as easy as pointing as clicking. And because the profiles are portable, users can switch from failover servers to spare servers as needed, which means users won’t need as many spare servers. In addition, the redundant switch and V8000 architectures add resiliency, which saves time and money in case of a hardware failure.

3Leaf’s approach differs from that of other I/O virtualization providers, says Reiner, because 3Leaf creates an open architecture leveraging commodity parts, including QLogic and Emulex HBAs to ease integration and certification. Plug-in cards and drivers also are commodity, which lets the company enable new features as they become available just by updating the drivers. By using commodity processors in the V8000 and supporting commercial switches, 3Leaf can follow the technology curves in these areas, as well, says Reiner. The real added value comes from the software, so being able to take advantage of hardware advances without heavy investment is a good thing. 3Leaf’s virtual HBAs and NICs also have adjustable QoA built in, so users have guaranteed bandwidth service levels.

Summing up I/O virtualization, George Crump, founder and president of analyst firm Storage Switzerland, said, “What it does today is allow you to solve that same sort of virtualization theme. It allows you to aggregate and better utilize existing network pipes for both storage and messaging … so you can reduce overall port count in the enterprise and reduce, in some cases, HBA count and things of that nature.” Crump added that while I/O virtualization is appealing mainly to large enterprises, it will not near the appeal of 3Leaf’s virtual compute and memory solution once that becomes available.

Completing the Virtual Compute Environment

Crump says the advanced virtualization adopters with whom he has spoken are crying for something like 3Leaf’s Virtual Compute Environment (VCE), often telling him they “need to get out of the box. VMware was great, it gave me a lot of good ideas, but I’m essentially landlocked in the sheet metal [of the individual server].” They want greater flexibility, the ability to scale when necessary, and the ability to pull in idle processors at peak times, said Crump, and these capabilities are not too common outside of grid environments.

Announced in early April, 3Leaf will complete the VCE strategy and form truly dynamic datacenters with its “game-changing” virtual compute and memory server. The solution enables “expandable” servers that can share memory and compute resources across physical boundaries thanks to 3Leaf’s one-of-a-kind hardware-based scheme, says Reiner. “It’s as is this is one large expandable [and] contractable server,” he added.

The solution allows for QoS management via built-in policies for failover, time-based events and resource monitoring, and it supports parallel processing, as well. For this reason, Reiner noted that 3Leaf expandable servers could be useful in some grid or cloud computing environment. However, understanding that some businesses require real-time performance management beyond 3Leaf’s scope, Reiner says the VCE solution is designed to handle external policy engines, and actually includes a third-party interface.

Taking advantage of AMD’s coherent HyperTransport and Intel’s QuickPath Interconnect technologies paired with 3Leaf’s proprietary ASIC, Reiner says 3Leaf’s expandable server technology will address the entire x86 market. Processors are connected to the 3Leaf proprietary ASIC, which then connects to a commodity 10 GbE or InfiniBand switch fabric. The silicon-based approach, says Reiner, will offer higher agility, scalability and utilization than software-based approaches, while — like the V8000 — drastically cutting CAPEX and OPEX.

“We’re not trying to make big SMP systems,” explained Reiner. “What we’re trying to do is deliver scalability and flexibility and agility into the x86 market. So within the enterprise datacenter, you can add resources and de-allocate resources as required to address changing business needs.”

But 3Leaf’s solution conflict with current VMware or XenServer environments, right? Not so, says Crump, who believes 3Leaf’s hardware-based, cross-machine VCE is more complementary to existing server virtualization technologies than anything else — especially in datacenters with high numbers of both physical and virtual servers. “If you have a few ESX servers and, for the most part, they’re solving all your problems, it’s doubtful you’ll need virtual compute,” he explained. “But if you [have] 50 to 100 physical servers with a fair amount [of them] that aren’t virtualized, that’s where that will appeal more.”

3Leaf’s virtual compute and memory solution will be available for AMD environments in the first half of 2009, and the Intel version will hit the market in 2010.

It Pays to Virtualize Everything

According to Reiner, 3Leaf’s goal is to tackle head-on the pressures to decrease skyrocketing operational expenditures under control while still addressing the ever-more-prevalent demands for increased flexibility and agility. He cites an IDC statistic stating that overprovisioning wastes $130 billion per year to justify 3Leaf’s focus on increasing utilization. “On the operation side,” he says, “there is a clear need to bring … expenditures under control.”

That 3Leaf meets it goal to cut costs is evidenced in its approximately dozen proof-of-concept evaluations with Fortune 100 customers. Ranging from e-commerce to social networking to financial services, Reiner says (as noted above) these evaluations have saved customers and average of 50 percent in both CAPEX and OPEX. “We are coming in with something that is very different and unique, and really solves a lot of the problems that have been around for a long time and never really have been addressed,” says Reiner.

As pressures to drive down costs while still increasing performance continue to mount, the temptation to “virtualize everything” will grow correspondingly. According to Crump, this trend probably will reach its pinnacle in two to four years after a gradual, slow roll. Server virtualization like VMware is only the first “baby steps,” he explained, and the next wave will encompass infrastructure and I/O virtualization, with compute and memory following behind. The reason for the gradual approach, he believes, is that because IT departments are always busy with a laundry list of tasks, problems generally are not addressed until they absolutely have to be. For example, he noted, the “green IT” craze did not start because everyone suddenly became environmentalists, but rather because they couldn’t get any more power. And the same thing is happening with virtualization.

“You can only ask IT to do more with less so many times before it becomes abominable,” says Crump. “Really, the only way to get there now in some of these environments is to virtualize everything, because the needs of the business have not slowed down – they’ve probably increased. The only way to be able to be responsive to those needs is to have a virtual environment at all layers – server, compute, storage, backup, network infrastructure [and] I/O infrastructure can all be virtualized and moved relatively quickly depending on today’s needs of the business.”

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!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues 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 power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion 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…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen 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…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. 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…

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…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a 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…

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…

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…

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…

Leading Solution Providers

Contributors

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…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t 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…

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…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

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…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire