Quantum Roundup: IBM Goes to School, Delft Tackles Networking, Rigetti Updates

By John Russell

September 5, 2019

IBM today announced a new open source quantum ‘textbook’, a series of quantum education videos, and plans to expand its nascent quantum hackathon program. Last month Delft University researchers described a potential quantum networking stack with a detailed proposal for a link layer protocol – all necessary for creating a quantum internet. Meanwhile Rigetti Computing is taking advantage of the one year anniversary of its cloud platform launch and announcement of a $1M prize for achieving quantum advantage to update media on progress.

While not blockbusters the activities reflect the steady pace of advance in quantum research. We’ll start with IBM but first mention that there’s been no winner yet in the Rigetti challenge (not really unexpected) and that developing a real quantum networking stack will be a big deal but remains far off. The Delft work is an important step forward.

IBM Goes Back to School

IBM’s latest offerings, announced in a blog by Jay Gambetta, IBM Fellow and quantum scientist, and Abraham Asfaw, IBM global lead of quantum education, are largely aimed at academia and intended to further seed growth in the quantum developer community.

Indeed all of the quantum computing hopefuls are seeking ways to spur ecosystem growth and IBM has been at the forefront. The IBM Q Experience was the first quantum computer accessed via the cloud (2016) and has gained 155,000 registered users since. Its software stack/developer kit, Qiskit, has more than 235,000 downloads and IBM Q users have already run over 28 million experiments and simulations. Notably, more than 190 papers have been written by non-IBMers related to quantum computing using the IBM Q systems.

Here’s a snapshot of the new resources:

  • Open-Source TextbookExplores quantum computing through practical problem sets run on real quantum systems, helping university students connect theory to practice.
  • Coding With QiskitAn approachable video series about quantum software that visually engages students through the hardest part of developing new language skills—the beginning! Students can learn quantum programming basics at their own pace with host, IBM Q developer advocate Abe Asfaw.
  • University Hackathon Partnership Program: Universities can partner with our teams to host a hands-on, collaborative Qiskit experience for eager quantum computing students.
  • New IBM Q Experience Systems: new quantum devices available over the cloud offer IBM Q Experience for Researcher program participants, and IBM Q Network organizations a scheduling feature for their experiments, educational demonstrations and

Bob Sutor, IBM VP of IBM Q Strategy and Ecosystem, briefed HPCwire on the new educational resources.

“The open source textbook (Learn Quantum Computation Using Qiskit) is going to be available to anybody,” said Sutor. “It’s licensed under the Apache to license, and it will evolve over time to explain the principles of quantum computing, with actual code that runs systems. So we’re introducing this now.

“For the videos, we have a very dynamic young researcher (Abraham Asfaw) who he is doing them, and they’re prepared by a larger team. He is a relatively new Ph.D. out of Princeton, and in a very informal but absolutely technically correct way, he’s introducing how you get up and running. The plan is there to introduce these videos about once a week until we cover the core material.”

Hackathons, of course, are nothing new but ramping up a Hackathon program for quantum computing is interesting. It makes quantum computing look more and more mainstream, at least given QC’s early stage. Although not mentioned in the blogpost, Sutor said IBM also plans to expand its quantum camps program aimed at graduate students.

“In a somewhat more formal away we’re going to continue with camps. These are typically three- or four-day workshops. We did one in in Vermont earlier this year and we have three coming up, one in Europe in Zurich, in a couple of weeks, another in Tokyo in mid-November, and one in Johannesburg in December. They are invitation only in the sense that we want graduate students who are recommended by professors,” said Sutor.

On the order of 22 universities are already affiliated with the IBM Q Network in one or another way according to Sutor who is hoping the outreach efforts send those numbers much higher. “We would love to quadruple them,” he said.

While today’s news focused on academia IBM’s broad goal is commercialization of QC. “At the core, quantum computing is a commercial program for IBM and everything we’ve done is focused on getting to quantum advantage as quickly as possible. This means having the hardware and the software and the systems and the people, writing the algorithms, working on the use cases, fundamentally, just understanding quantum computing.”

Delft Researchers Propose a Quantum Network Protocol Stack

Delft University Researchers, led by Stephanie Wehner, presented a rich paper (A Link Layer Protocol for Quantum Networks) at the recent ACM SIGCOMM conference examining the challenges facing building a quantum communication network stack. There has been lost of activity around quantum-based communications.

So far the efforts have been relatively modest but distances are growing. In their paper, the researchers note short-lived entanglement has been produced probabilistically over “short distances (100 km) on the ground by sending photons over standard telecom fiber as well as from space over 1203 km from a satellite.” These point-to-point communication successes are impressive but can only prepare and measure single qubits and cannot by be concatenated to allow the transmission of qubits over longer distances.

Among the technology needs for effective quantum networking is a robust networking stack whose protocols are specifically designed for quantum.

Here’s a brief excerpt from the paper:

“We take the first step from a physics experiment to a quantum internet system. We propose a functional allocation of a quantum network stack, and construct the first physical and link layer protocols that turn ad-hoc physics experiments producing heralded entanglement between quantum processors into a well-defined and robust service. This lays the groundwork for designing and implementing scalable control and application protocols in platform-independent software.

“To design our protocol, we identify use cases, as well as fundamental and technological design considerations of quantum network hardware, illustrated by considering the state-of-the-art quantum processor platform available to us (Nitrogen-Vacancy (NV) centers in diamond.”

Not surprisingly classical network protocols don’t transfer to quantum. Among the many challenges in any quantum networking scheme is determining whether or not a pair of qubits travelling through the network are entangled. There are approaches to doing this. Wehner and her colleagues discuss an approach called heralded entanglement generation that generates a heralding signal that can be sent between node to confirm entanglement. Their work is fairly detailed examination of issues, use cases, and in the case of the link layer, a specific proposal to build upon.

Here’s outline of the proposed stack elements excerpted from their paper:

  • “Physical layer. This layer is realized by the actual quantum hardware devices and physical connections such as fibers. We take the physical layer to contain no decision-making elements and keep no state about the production of entanglement (or the transmissions of qubits). The hardware at the physical layer is responsible for timing synchronization and other synchronization, such as laser phase stabilization [47], required to make attempts to produce heralded entanglement.
  • “Link layeris [used] then to turn the physical layer making entanglement attempts into a robust entanglement generation service, that can produce entanglement between controllable quantum nodes connected by an (chain of) automated quantum node. Requests can be made by higher layers to the link layer to produce entanglement, where robust means that the link layer endows the physical system with additional guarantees: a request for entanglement generation will (eventually) be fulfilled or result in a time-out.
  • “Network layer is responsible for producing long-distance entanglement between nodes that are neither connected directly, nor connected by a chain of automated quantum nodes at the physical layer. This may be achieved by means of entanglement swapping, using the link layer to generate entanglement between neighboring controllable nodes. In addition, it contains an entanglement manager that keeps track of entanglement in the network, and which may choose to pre- generate entanglement to service later requests from higher layers.
  • “Transport layer takes responsibility for transmitting qubits deterministically (e.g. using teleportation). One may question why this warrants a separate layer, rather than a library. Use of a dedicated layer allows two nodes to pre-share entanglement that is used as applications of the system demand it. Here, entanglement is not assigned to one specific application. This potentially increases the throughput of qubit transmission via teleportation, as teleportation requires no additional connection negotiation, but only forward communication from a sender to the receiver.”

There’s a lot to unpack here and the paper is best read directly. In an account of the work posted on Delft web site Wehner said, “Currently, qubits cannot be kept in memory for very long. This means control decisions on what to do with them need to be taken very quickly. By creating this link layer protocol, we have overcome obstacles presented by some very demanding physics.”

Recognizing a great deal of work still needs to be done, the researchers nevertheless sound a positive note in their conclusion, “Our top down inventory of design requirements, combined with a bottom up approach based on actual quantum hardware allowed us to take quantum networks a step further on the long path towards their large-scale realization. Our work paves the way towards the next step, a robust network layer control protocol. The link layer may now be used as a robust service without detailed knowledge of the physics of the devices.”

Rigetti Doubles Down on Leveraging Hybrid Approach

Almost exactly one year ago Rigetti Computing introduced its Quantum Cloud Service (QCS) platform which emphasize optimizing a hybrid quantum-classical approach in the race to achieve quantum advantage. QA is the idea that at some point quantum computers will perform some applications sufficiently better than classical systems to warrant switching. Rigetti also introduced $1 million prize for the first person to achieve QA using its platform. (see HPCwire article, Rigetti (and Others) Pursuit of Quantum Advantage)

This week Rigetti’s SVP of Engineering and Product, David Rivas, spoke with HPCwire to provide an update. Rigetti emphasizes its commitment to optimizing hybrid quantum-classical computing and describes itself as a “full stack” company meaning it controls all critical elements, and among other things intentional optimizes for the hybrid quantum-classical style of quantum computing likely to succeed near-term.

Hybrid is a word frequently bandied about in connection with quantum computing and really it has two meanings.

  • In practical terms you can’t do quantum computing without classical computing. Think of classical computing as a kind of envelope around all quantum systems that is necessary to get data in out of the system and to control the system. Also, since the quantum machine’s answers are probabilistic, classical systems are needed to convert output into definite answers. In this sense, all quantum computers are ‘hybrid systems’ and require classical systems for basic tasks.
  • The second meaning has to do with how we process quantum algorithms. Trying to run these algorithms to solve large problems on today’s low qubit-count machines is impractically resource intensive (for example, extra qubits for error or memory). By breaking the algorithms into pieces, it’s possible to run the portions of the algorithm which require quantum properties on the quantum processor – or put another way make the best use of quantum properties – and to run other portions on a classical computer. Since quantum answers are probabilistic, the loop between the quantum processor and the classical system is typically repeated many times before eventually converging on a solution.

Clever design and use of these hybrid algorithms are what is expected to enable today’s imperfect, low qubit-count quantum machines to do useful work.

“Near term, it’s going to be those kinds of algorithms that we think are going to take us towards quantum advantage, ones where the quantum processor is performing a task that is uniquely difficult to do in a classical environment, but which the quantum processor is itself uniquely capable doing very quickly. But the classical and quantum components are deeply intertwined,” said Rivas.

Maximizing the efficient running of these hybrid algorithms is a Rigetti emphasis. For example, latencies in communication between the quantum processor and classical system can be problematic. This is especially so when accessing a quantum system by a web portal where the two systems are remote from each other as was the case initially for users accessing to Rigetti’s quantum system via an API. That changed when QCS was launched.

“We decided there were two major improvements we could make. The most dramatic is colocation. We put the classical resources very close, both physically and network connectivity, to the quantum resource. By doing this, we decrease the latencies by one or two orders of magnitude. What would take minutes takes seconds or less,” said Rivas.

It also turns out that for many important quantum algorithm – particularly the Variational-Quantum-Eigensolver (VQE) – tuning the parameters is important as the calculation proceeds.

Rigetti added automated support for parametrization. “You have an inner loop, a quantum circuit that you want to run, and then you have an optimization [tuning parameters] that is performed classically. The innovation here was to support the notion of parameterizing circuits [in a way] that you remove the compilation step and the transmission of the circuit stuff from the actual communication between the classical and the quantum program,” said Rivas.

A third advance involves systems reset.  Generally, after quantum processor runs a circuit it needs to be reset to a known state. The traditional way of doing this is to sort of just wait for the processor settle. Rigetti implemented an active reset which speeds the process – “The technical detail here is we run a very small circuit a number of times to set that cubits into a known state, but we do it efficiently and at the control level.”

Rivas said, “The combination of these things, the colocation, the parameterizing, and the active reset has resulted in a 30X and 50X performance increases for these kinds of applications.”

Progress in build bigger quantum processors has been somewhat slower. The goal cited at the time of the cloud platform launch was to reach 128 qubit processors in roughly a year.

“We’re certainly on a path towards 100-plus qubit, but it probably won’t appear that by the end of this year. We have successfully running now, 32 qubits in our lab and close to production. The important that we’re doing there is we’re driving our noise metrics very, very low. I don’t think we’re yet announcing numbers associated with that, but the intention here is to is to focus on both the number of cubits and the overall specifications associated with qubit,” said Rivas.

Feature image: Using the link layer protocol, higher-layer software can request the creation of entanglement without needing to know which quantum hardware system is in the box.
Image credit QuTech/Scixel.

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