Quantum Bits: D-Wave and VW; Google Quantum Lab; IBM Expands Access

By John Russell

March 21, 2017

For a technology that’s usually characterized as far off and in a distant galaxy, quantum computing has been steadily picking up steam. Just how close real-world applications are depends on whom you talk to and for what kinds of applications. Los Alamos National Lab, for example, has an active application development effort for its D-Wave system and LANL researcher Susan Mniszewski and colleagues have made progress on using the D-Wave machine for aspects of quantum molecular dynamics (QMD) simulations.

At CeBIT this week D-Wave and Volkswagen will discuss their pilot project to monitor and control taxi traffic in Beijing using a hybrid HPC-quantum system – this is on the heels of recent customer upgrade news from D-Wave (more below). Last week IBM announced expanded access to its five-qubit cloud-based quantum developer platform. In early March, researchers from the Google Quantum AI Lab published an excellent commentary in Nature examining real-world opportunities, challenges and timeframes for quantum computing more broadly. Google is also considering making its homegrown quantum capability available through the cloud.

As an overview, the Google commentary provides a great snapshot, noting soberly that challenges such as the lack of solid error correction and the small size (number of qubits) in today’s machines – whether “universal” digital machines like IBM’s or “analog” adiabatic annealing machines like D-Wave’s – have prompted many observers to declare useful quantum computing is still a decade way. Not so fast, says Google.

“This conservative view of quantum computing gives the impression that investors will benefit only in the long term. We contend that short-term returns are possible with the small devices that will emerge within the next five years, even though these will lack full error correction…Heuristic ‘hybrid’ methods that blend quantum and classical approaches could be the foundation for powerful future applications. The recent success of neural networks in machine learning is a good example,” write Masoud Mohseni, Peter Read, and John Martinis (a 2017 HPCwire Person to Watch) and colleagues (Nature, March 8, “Commercialize early quantum technologies”)

The D-Wave/VW project is a good example of a hybrid approach (details to follow) but first here’s a brief summary of recent quantum computing news:

  • IBM released a new API and upgraded simulator for modeling circuits up to 20 qubits on its 5-qubit platform. It also announced plans for a software developer kit by mid-year for building “simple” quantum applications. So far, says IBM, its quantum cloud has attracted about 40,000 users, including, for example, the Massachusetts Institute of Technology, which used the cloud service for its online quantum information science course. IBM also noted heavy use of the service by Chinese researchers. (See HPCwire coverage, IBM Touts Hybrid Approach to Quantum Computing)
  • D-Wave has been actively extending its development ecosystem (qbsolv (D-wave) and qmasm (LANL, et al.) and says researchers have recently been able to simulate a 20,000 qubit system on 1,000-qubit machine using qbsolv (more below). After announcing a 2,000-quibit machine in the fall, the company has begun deploying them. The first will be for a new customer, Temporal Defense System, and another is planned for the Google/NASA/USRA partnership which has a 1,000-qubit machine now. D-wave also just announced Virginia Tech and the Hume Center will begin using D-Wave systems for work on defense and intelligence applications.
  • Google’s commentary declares: “We anticipate that, within a few years, well-controlled quantum systems may be able to perform certain tasks much faster than conventional computers based on CMOS (complementary metal oxide–semiconductor) technology. Here we highlight three commercially viable uses for early quantum-computing devices: quantum simulation, quantum-assisted optimization and quantum sampling. Faster computing speeds in these areas would be commercially advantageous in sectors from artificial intelligence to finance and health care.”
D-Wave 2000Q System

Clearly there is a lot going on even at this stage of quantum computing’s development. There’s also been a good deal of wrangling over just what is a quantum computer and the differences between IBM’s “universal” digital approach – essentially a machine able to do anything computers do now – and D-Wave’s adiabatic annealing approach, which is currently intended to solve specific classes of optimization problems.

“They are different kinds of machines. No one has a universal quantum computer now, so you have to look at each case individually for its particular strengths and weaknesses,” explained Martinis to HPCwire. “The D-wave has minimal quantum coherence (it loses the information exchanged between qubits quite quickly), but makes up for it by having many qubits.”

“The IBM machine is small, but the qubits have quantum coherence enough to do some standard quantum algorithms. Right now it is not powerful, as you can run quantum simulations on classical computers quite easily. But by adding qubits the power will scale up quickly. It has the architecture of a universal machine and has enough quantum coherence to behave like one for very small problems,” Martinis said.

Noteworthy, Google has developed 9-qubit devices that have 3-5x more coherence than IBM, according to Martinis, but they are not on the cloud yet. “We are ready to scale up now, and plan to have this year a ‘quantum supremacy’ device that has to be checked with a supercomputer. We are thinking of offering cloud also, but are more or less waiting until we have a hardware device that gives you more power than a classical simulation.”

Quantum supremacy as described in the Google commentary is a term coined by theoretical physicist John Preskill to describe “the ability of a quantum processor to perform, in a short time, a well-defined mathematical task that even the largest classical supercomputers (such as China’s Sunway TaihuLight) would be unable to complete within any reasonable time frame. We predict that, in a few years, an experiment achieving quantum supremacy will be performed.”

Bo Ewald

For the moment, D-Wave is the only vendor offering near-production machines versus research machines, said Bo Ewald, the company’s ever-cheerful evangelist. He quickly agrees though that at least for now there aren’t any production-ready applications. Developing a quantum tool/software ecosystem is a driving focus at D-wave. The LANL app dev work, though impressive, still represents proto-application development. Nevertheless the ecosystem of tools is growing quickly.

“We have defined a software architecture that has several layers starting at the quantum machine instruction layer where if you want to program in machine language you are certainly welcome to do that; that is kind of the way people had to do it in the early days,” said Ewald.

“The next layer up is if you want to be able to create quantum machine instructions from C or C++ or Python. We have now libraries that run on host machines, regular HPC machines, so you can use those languages to generate programs that run on the D-Wave machine but the challenge that we have faced, that customers have faced, is that our machines had 500 qubits or 1,000 qubits and now 2,000; we know there are problems that are going to consume many more qubits than that,” he said.

For D-Wave systems, qbsolv helps address this problem. It allows a meta-description of the machine and the problem you want to solve as quadratic unconstrained binary optimization or QUBO. It’s an intermediate representation. D-Wave then extended this capability to what it calls virtual QUBOs likening it to virtual memory.

“You can create QUBOs or representations of problems which are much larger than the machine itself and then using combined classical computer and quantum computer techniques we could partition the problem and solve them in chunks and then kind of glue them back together after we solved the D-Wave part. We’ve done that now with the 1,000-qubit machine and run problems that have the equivalent of 20,000 qubits,” said Ewald, adding the new 2,000-qubit machines will handle problems of even greater size using this capability.

At LANL, researcher Scott Pakin has developed another tool – a quantum macro assembler for D-Wave systems (QMASM). Ewald said part of the goal of Pakin’s work was to determine, “if you could map gates onto the machine even though we are not a universal or a gate model. You can in fact model gates on our machine and he has started to [create] a library of gates (or gates, and gates, nand gates) and you can assemble those to become macros.”

Pakin said,My personal research interest has been in making the D-Wave easier to program. I’ve recently built something really nifty on top of QMASM: edif2qmasm, which is my answer to the question: Can one write classical-style code and run it on the D-Wave?

“For many difficult computational problems, solution verification is simple and fast. The idea behind edif2qmasm is that one can write an ordinary(-ish) program that reports if a proposed solution to a problem is in fact valid. This gets compiled for the D-Wave then run _backwards_, giving it ‘true’ for the proposed solution being valid and getting back a solution to the difficult computational problem.”

Pakin noted there are many examples on github to provide a feel for the power of this tool.

“For example, mult.v is a simple, one-line multiplier. Run it backwards, and it factors a number, which underlies modern data decryption. In a dozen or so lines of code, circsat.v evaluates a Boolean circuit. Run it backwards, and it tells you what inputs lead to an output of “true”, which used in areas of artificial intelligence, circuit design, and automatic theorem proving. map-color.v reports if a map is correctly colored with four colors such that no two adjacent regions have the same color. Run it backwards, and it _finds_ such a coloring.

“Although current-generation D-Wave systems are too limited to apply this approach to substantial problems, the trends in system scale and engineering precision indicate that some day we should be able to perform real work on this sort of system. And with the help of tools like edif2qmasm, programmers won’t need an advanced degree to figure out how to write code for it,” he explained.

The D-Wave/VW collaboration, just a year or so old, is one of the more interesting quantum computing proof-of-concept efforts because it tackles an optimization problem of the kind that is widespread in everyday life. As described by Ewald, VW CIO Martin Hoffman was making his yearly swing through Silicon Valley and stopped in at D-Wave and talk turned to the many optimization challenges big automakers face, such as supply logistics, vehicle delivery, and various machine learning tasks and doing a D-Wave project around one of them. Instead, said Ewald, VW eventually settled on a more driver-facing problem.

It turns out there are about 10,000 taxis in Beijing, said Ewald. Each has a GPS device and their positions are recorded every five seconds. Traffic congestion, of course, is a huge problem in Beijing. The idea was to explore if it was possible to create an application running on both traditional computer resources and D-Wave to help monitor and guide taxi movement more quickly and effectively.

“Ten thousand taxis on all of the streets in Beijing is way too big for our machine at this point, but they came to this same idea we talked about with qbsolve where you partition problems,” said Ewald. “On the traditional machines VW created a map and grid and subdivided the grid into quadrants and would find the quadrant that was the most red.” That’s red as in long cab waits.

The problem quadrant was then sent to D-Wave to be solved. “We would optimize the flow, basically minimize the wait time for all of the taxis within the quadrant, send that [solution] back to the traditional machine which would then send us the next most red, and we would try to turn it green,” said Ewald.

According to Ewald, VW was able to relatively create the “hybrid” solutions quickly and “get what they say are pretty good results.” They have talked about then being able to extend this project to predict where traffic jams are going to be and give people perhaps 45 minute warnings that there’s the potential for a traffic jam at such and such intersection. The two companies have a press conference planned this week at CeBIT to showcase the project.

It’s good to emphasize that the VW/D-wave exercise is developmental – what Ewald labels as a proto application: “But just the fact that they were able to get it running is a great step forward in many ways in that we believe our machine will be used side by side with existing machines, much like GPUs were used in the early days on graphics. In this case VW has demonstrated quite clearly how our machine, our QPU if you will, can be used in helping accelerate the work being done on a traditional HPC machines.”

Image art, chip diagram: D-Wave

Subscribe to HPCwire's Weekly Update!

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

Graphcore Introduces Larger-Than-Ever IPU-Based Pods

October 22, 2021

Graphcore and its “Intelligent Processing Units” (IPUs) emerged from stealth in 2016 and launched its second-generation IPU in 2020. While the company has also launched its IPUs in a variety of form factors over the Read more…

Quantum Chemistry Project to Be Among the First on EuroHPC’s LUMI System

October 22, 2021

Finland’s CSC has just installed the first module of LUMI, a 550-peak petaflops system supported by the European Union’s EuroHPC Joint Undertaking. While LUMI -- pictured in the header -- isn’t slated to complete i Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat for that approach), announced it was expanding into gate-based Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the power of Fugaku, the current top-ranked supercomputer in the wor Read more…

AWS Solution Channel

Royalty-free stock illustration ID: 537899029

Running GROMACS on GPU instances

Comparing the performance of real applications across different Amazon Elastic Compute Cloud (Amazon EC2) instance types is the best way we’ve found for finding optimal configurations for HPC applications here at AWS. Read more…

Faster Optical Switch that Operates at ‘Room Temp’ Developed by IBM, Skolkovo Researchers

October 19, 2021

Optical switching technology holds great promise for many applications but hot operating temperatures have been a key obstacle slowing progress. Now, a new optical switching device that can operate at room temperatures a Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the powe Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Quantum Workforce – NSTC Report Highlights Need for International Talent

October 13, 2021

Attracting and training the needed quantum workforce to fuel the ongoing quantum information sciences (QIS) revolution is a hot topic these days. Last week, the U.S. National Science and Technology Council issued a report – The Role of International Talent in Quantum Information Science... Read more…

Eni Returns to HPE for ‘HPC4’ Refresh via GreenLake

October 13, 2021

Italian energy company Eni is upgrading its HPC4 system with new gear from HPE that will be installed in Eni’s Green Data Center in Ferrera Erbognone (a provi Read more…

The Blueprint for the National Strategic Computing Reserve

October 12, 2021

Over the last year, the HPC community has been buzzing with the possibility of a National Strategic Computing Reserve (NSCR). An in-utero brainchild of the COVID-19 High-Performance Computing Consortium, an NSCR would serve as a Merchant Marine for urgent computing... Read more…

UCLA Researchers Report Largest Chiplet Design and Early Prototyping

October 12, 2021

What’s the best path forward for large-scale chip/system integration? Good question. Cerebras has set a high bar with its wafer scale engine 2 (WSE-2); it has 2.6 trillion transistors, including 850,000 cores, and was fabricated using TSMC’s 7nm process on a roughly 8” x 8” silicon footprint. Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

US Closes in on Exascale: Frontier Installation Is Underway

September 29, 2021

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, held by Zoom this week (Sept. 29-30), it was revealed that the Frontier supercomputer is currently being installed at Oak Ridge National Laboratory in Oak Ridge, Tenn. The staff at the Oak Ridge Leadership... Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer... Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Leading Solution Providers

Contributors

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-ap Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make i Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

Quantum Computer Market Headed to $830M in 2024

September 13, 2021

What is one to make of the quantum computing market? Energized (lots of funding) but still chaotic and advancing in unpredictable ways (e.g. competing qubit tec Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire