Dell Builds Tech for Quantum Computers to Interface with Conventional Systems – HPCwire
Since 1987 – Covering the Fastest Computers in the World and the People Who Run Them
Since 1987 – Covering the Fastest Computers in the World and the People Who Run Them
November 14, 2022
Dell is creating the building blocks to assimilate quantum computers into conventional IT infrastructures as it opens datacenters to new types of accelerated computers.
The server-maker has created a blueprint for conventional server infrastructure to interface with the unique demands of quantum systems, which are considered significantly faster and solve problems in a different way.
The hybrid classical-quantum computing model includes an adjusted hierarchy with conventional servers acting as hosts and managers of attached quantum computers, which compute differently.
The blueprint also includes a conventional computing layer so relevant results from quantum calculations – which aren’t straightforward answers – can be sent back to users within an IT infrastructure.
“We are already seeing those areas that we can optimize and make that intelligent interplay between the classical infrastructure and the quantum infrastructure even more powerful in order to do things that we even haven’t considered yet,” Ken Durazzo, vice president of research at Dell told HPCwire.
The hybrid computing blueprint, which includes hardware and software, is built around PowerEdge servers and quantum computing technologies from IonQ. The companies announced a hybrid system, Dell Quantum Computing Solution, based on the design at the SC22 conference. The system will be available in the U.S. and Canada.
Quantum computers are widely seen as one way to advance computing beyond today’s PCs and servers. Quantum computers perform significantly more calculations in parallel, which provides a vast increase in processing power compared to conventional servers.
“As we’ve seen on this journey with quantum, we realized that it’s not quantum computing as the future, it’s classical and quantum working together,” Durazzo said.
Durazzo said the top-end GPUs are being taxed by applications such as computational fluid dynamics and chemistry. Dell’s hybrid model involves running applications on the classical infrastructure, and algorithms from those applications being accelerated via quantum computing.
“We identified very early on … that there are clear roles for the classical infrastructure and then for the quantum infrastructure,” Durazzo said.
Dell’s hybrid model accounts for the vast amounts of data ingested by quantum computers, which runs calculations at breakneck speeds. The conventional infrastructure creates the conditions for quantum calculations before data is fed into quantum processors, which has its own quirks in the way it calculates and delivers results.
For example, conventional systems spit out answers resulting from calculations on circuits like CPUs and GPUs. By comparison, quantum computers don’t deliver answers, but instead send back probabilistic results. Those could be any number of results, which are delivered in some level of stack rank order.
“When the algorithm gets sent to the QPU [quantum processing unit], the QPU sends back results of the computation. They don’t send back an answer, they send back results of the computation. Those results will be taken into the classical system again, and will be formulated into a result that the end user could actually see coming from the application,” Durazzo said.
The foundation for the hybrid model is a set up of the server, networking and other components, where the program actually runs and initiates high-speed data transfers to quantum systems. The data has to pass through an intermediary host processor, which is more of an instruction layer for conventional systems to interface with the quantum systems.
“What the host processor in this case is actually doing is taking that circuit, compiling it into QASM, or the quantum assembly language, and then taking that to program the QPU. That host processor is an intermediary step in order to program the QPU and to provide the algorithm for the QPU to actually exercise,” Durazzo said.
Over time, the host processor will be able to do “very intelligent scheduling operations,” Durazzo said.
For example, instead of sending an entire quantum computing task to a physical quantum processor, the host processor will be able to decompose a quantum task and bifurcate its execution between a physical quantum hardware and software-based quantum simulator on GPUs.
“We’re going see more and more models of that interplay between classical simulation and actual physical quantum systems, where the intelligent orchestration and the ability to understand where to best run these particular functions or operations is going to become more critical,” Durazzo said, adding “We’re well positioned for that world as that emerges.”
A classical system will then try to make sense of the complicated quantum output and formulate a concrete answer from an application for end users. The classical system could be equipped with accelerators like GPUs, which could use AI techniques to crunch the quantum output to generate relevant answers. The hybrid model uses standard software stacks that includes Kubernetes for orchestration and QISKIT for quantum applications.
“We also designed it to enable an applications developer to write the application or the algorithm one time and then run it anywhere they need to,” such as on a physical quantum circuit or a quantum emulator on a GPU, Durazzo said.
Access to quantum computers are already available through Amazon and Google clouds. Dell hopes its managed services offering will make it easier for companies to smoothly transition to quantum systems, which could be a daunting task.
Dell’s partnership with IonQ is just the start of its quantum journey, and many partnerships with quantum computer providers could be on tap, Durazzo said.
Each quantum computer vendor offers its own type of qubit, and customers are eager to test different systems to find the right hardware that fits their infrastructure. IonQ offers the trapped-ion qubit, Google and IBM are working on superconducting qubits, and D-Wave offers a quantum annealing system for optimization tasks.
“In this case, we’re selling with IonQ, but we can easily add support … [from] multiple vendors to allow for investment protection as well as to allow for experimentation on these different quantum systems because they all behave just a little bit different,” Durazzo said.
More Off The Wire
Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!
November 14, 2022
Nvidia’s H100 GPU, the flagship of its Hopper architecture, has debuted on the Top500 and Green500 lists at SC22. The new GPU appears in the relatively small Lenovo-built Henri system, which also features Intel’s Xeo Read more…
November 14, 2022
The 60th edition of the Top500 list, revealed today at SC22 in Dallas, Texas, showcases many of the same systems as the previous installment, with Frontier still out in front as the first official Linpack exascaler, cloc Read more…
November 14, 2022
MLCommons last week issued its third annual set of MLPerf HPC (v2.0) results intended to showcase the performance of larger systems when training more rigorous scientific models. The large size of systems participating i Read more…
November 14, 2022
Dell is creating the building blocks to assimilate quantum computers into conventional IT infrastructures as it opens datacenters to new types of accelerated computers. The server-maker has created a blueprint for con Read more…
November 14, 2022
Over the past months, Nvidia has put a spotlight on its OVX hardware – purpose-built systems aimed at its Omniverse digital twins platform. Now, at SC22, Nvidia is launching new Omniverse capabilities aimed at leveragi Read more…
Today we announced the AWS Impact Computing Project at the Harvard Data Science Initiative (HDSI) to identify potential solutions that can improve the lives of humans, other species, and natural ecosystems. Read more…
Contributors: John W. Schmotzer, Automotive Ecosystem Business Development Manager, NVIDIA and Guy Bursell, Business Strategy Lead, Manufacturing & Supply Chain at Microsoft
The automotive industry continues to adopt artificial intelligence (AI) at an increasingly rapid pace to stay abreast of numerous production challenges and market opportunities. Read more…
November 14, 2022
Cerebras is putting down stakes to be a player in the AI cloud computing with a supercomputer called Andromeda, which achieves over an exaflops of “AI performance.” The company called Andromeda one of the fastest AI s Read more…
November 14, 2022
Nvidia’s H100 GPU, the flagship of its Hopper architecture, has debuted on the Top500 and Green500 lists at SC22. The new GPU appears in the relatively small Read more…
November 14, 2022
The 60th edition of the Top500 list, revealed today at SC22 in Dallas, Texas, showcases many of the same systems as the previous installment, with Frontier stil Read more…
November 14, 2022
MLCommons last week issued its third annual set of MLPerf HPC (v2.0) results intended to showcase the performance of larger systems when training more rigorous Read more…
November 14, 2022
Dell is creating the building blocks to assimilate quantum computers into conventional IT infrastructures as it opens datacenters to new types of accelerated co Read more…
November 14, 2022
Over the past months, Nvidia has put a spotlight on its OVX hardware – purpose-built systems aimed at its Omniverse digital twins platform. Now, at SC22, Nvid Read more…
November 14, 2022
Cerebras is putting down stakes to be a player in the AI cloud computing with a supercomputer called Andromeda, which achieves over an exaflops of “AI performan Read more…
November 12, 2022
Chipmakers regularly indulge in a game of brinkmanship, with an example being Intel and AMD trying to upstage one another with server chip launches this week. But each of those companies are in different positions, with AMD playing its traditional role of a scrappy underdog trying to unseat the behemoth Intel… Read more…
November 10, 2022
AMD’s fourth-generation Epyc processor line has arrived, starting with the “general-purpose” architecture, called “Genoa,” the successor to third-gen Eypc Milan, which debuted in March of last year. At a launch event held today in San Francisco, AMD announced the general availability of the latest Epyc CPUs with up to 96 TSMC 5nm Zen 4 cores… Read more…
September 23, 2022
Nvidia is not interested in bringing software support to its GPUs for the RISC-V architecture despite being an early adopter of the open-source technology in its GPU controllers. Nvidia has no plans to add RISC-V support for CUDA, which is the proprietary GPU software platform, a company representative… Read more…
August 30, 2022
It is perhaps not surprising that the big cloud providers – a poor term really – have jumped into quantum computing. Amazon, Microsoft Azure, Google, and th Read more…
July 19, 2022
The U.S. Senate on Tuesday passed a major hurdle that will open up close to $52 billion in grants for the semiconductor industry to boost manufacturing, supply chain and research and development. U.S. senators voted 64-34 in favor of advancing the CHIPS Act, which sets the stage for the final consideration… Read more…
August 22, 2022
Amid the high-performance GPU turf tussle between AMD and Nvidia (and soon, Intel), a new, China-based player is emerging: Biren Technology, founded in 2019 and headquartered in Shanghai. At Hot Chips 34, Biren co-founder and president Lingjie Xu and Biren CTO Mike Hong took the (virtual) stage to detail the company’s inaugural product: the Biren BR100 general-purpose GPU (GPGPU). “It is my honor to present… Read more…
August 16, 2022
Tesla has revealed that its biggest in-house AI supercomputer – which we wrote about last year – now has a total of 7,360 A100 GPUs, a nearly 28 percent uplift from its previous total of 5,760 GPUs. That’s enough GPU oomph for a top seven spot on the Top500, although the tech company best known for its electric vehicles has not publicly benchmarked the system. If it had, it would… Read more…
November 12, 2022
Chipmakers regularly indulge in a game of brinkmanship, with an example being Intel and AMD trying to upstage one another with server chip launches this week. But each of those companies are in different positions, with AMD playing its traditional role of a scrappy underdog trying to unseat the behemoth Intel… Read more…
October 12, 2022
Most talk about quantum computing today, at least in HPC circles, focuses on advancing technology and the hurdles that remain. There are plenty of the latter. F Read more…
August 2, 2022
The Universal Chiplet Interconnect Express (UCIe) consortium is moving ahead with its effort to standardize a universal interconnect at the package level. The c Read more…
September 2, 2022
Fusion, the nuclear reaction that powers the Sun and the stars, has incredible potential as a source of safe, carbon-free and essentially limitless energy. But Read more…
September 8, 2022
The steady maturation of MLCommons/MLPerf as an AI benchmarking tool was apparent in today’s release of MLPerf v2.1 Inference results. Twenty-one organization Read more…
August 3, 2022
After two-plus years of contentious debate, several different names, and final passage by the House (243-187) and Senate (64-33) last week, the Chips and Science Act will soon become law. Besides the $54.2 billion provided to boost US-based chip manufacturing, the act reshapes US science policy in meaningful ways. NSF’s proposed budget… Read more…
October 6, 2022
Intel is opening up its fabs for academic institutions so researchers can get their hands on physical versions of its chips, with the end goal of boosting semic Read more…
August 12, 2022
Courtesy of the schedule for the SC22 conference, we now have our first glimpse at the finalists for this year’s coveted Gordon Bell Prize. The Gordon Bell Pr Read more…
August 24, 2022
Fresh from finalizing its acquisitions of FPGA provider Xilinx (Feb. 2022) and DPU provider Pensando (May 2022) ), AMD previewed what it calls a 400 Gig Adaptive smartNIC SOC yesterday at Hot Chips. It is another contender in the increasingly crowded and blurry smartNIC/DPU space where distinguishing between the two isn’t always easy. The motivation for these device types… Read more…
August 11, 2022
A Google-led program to design and manufacture chips for free is becoming popular among researchers and computer enthusiasts. The search giant’s open silicon program is providing the tools for anyone to design chips, which then get manufactured. Google foots the entire bill, from a chip’s conception to delivery of the final product in a user’s hand. Google’s… Read more…
August 17, 2022
“It is my privilege to welcome you to the dedication of Frontier, the supercomputer that broke the exascale barrier.” That was the introduction by Oak Ridge National Laboratory Director Thomas Zacharia, at a small, public event on August 17 to officially dedicate the supercomputer, which in May became the first system to achieve over 1.0 exaflops of 64-bit performance on the… Read more…
© 2022 HPCwire. All Rights Reserved. A Tabor Communications Publication
HPCwire is a registered trademark of Tabor Communications, Inc. Use of this site is governed by our Terms of Use and Privacy Policy.
Reproduction in whole or in part in any form or medium without express written permission of Tabor Communications, Inc. is prohibited.