IBM Pitches Quantum Volume as Benchmarking Tool for Gate-based Quantum Computers

By John Russell

March 6, 2019

IBM this week announced it had achieved its highest Quantum Volume number to date at the American Physical Society (APS) March meeting being held in Boston. What’s Quantum Volume, you ask? Broadly, it’s a ‘holistic measure’ introduced by IBM in a paper last November that’s intended to characterize gate-based quantum computers, regardless of their underlying technology (semiconductor, ion trap, etc.), with a single number. IBM is urging wide adoption of QV by the quantum computing community.

The idea is interesting. The highest QV score so far is 16 which was attained by IBM’s fourth generation 20-qubit IBM Q System One; that’s double the QV of IBM’s 20-qubit IBM Q Network devices. You can see qubit count isn’t the determinant (but it is a factor). Many system-wide facets – gate error rates, decoherence times, qubit connectivity, operating software efficiency, and more – are effectively baked into the measure. In the paper, IBM likens QV to LINPACK for its ability to compare diverse systems.

IBM has laid out a roadmap in which it believes it can roughly double QVs every year. This rate of progress, argues IBM, will produce quantum advantage – which IBM defines as “a quantum computation is either hundreds or thousands of times faster than a classical computation, or needs a smaller fraction of the memory required by a classical computer, or makes something possible that simply isn’t possible now with a classical computer” – in the 2020s.

Addison Snell, CEO, Intersect360 Research noted, “Quantum volume is an interesting metric for tracking progress toward the ability to leverage quantum computing in ways that would be impractical for conventional supercomputers. With the different approaches to quantum computing, it is difficult to compare this achievement across the industry, but it is nevertheless a compelling statistic.”

There’s a lot to unpack here and it’s best done by reading the IBM paper, which isn’t overly long. Bob Sutor, VP, IBM Q Strategy and Ecosystem, and Sarah Sheldon, research staff at IBM T.J. Watson Research Center, briefed HPCwire on QV’s components, use, and relevance to the pursuit of quantum advantage. Before jumping into how Quantum Value is determined, Sutor’s comments on timing and what the magic QV number might be to achieve quantum advantage are interesting.

“We’re not going to go on record saying this or that particular QV number [will produce quantum advantage]. We have now educated hunches based on the different paths that people are taking, that people are taking for chemistry, for AI explorations, for some of the Monte Carlo simulations, and frankly the QV number may be different and probably will be different for each of those. We are certainly on record as saying in the 2020s and we hope in 3-to-5 years,” said Sutor.

The APS meeting served as a broad launchpad for QV with IBM making several presentations on various quantum topics while also seeking to stimulate conversation and urge adoption of QV within the gate-based quantum computing crowd. IBM issued a press release, a more technical blog with data points, and continued promoting the original paper (Validating quantum computers using randomized model circuits) which is freely downloadable. Rigetti has reportedly implemented QV. Noteworthy, QV is not meant for use with adiabatic annealing quantum systems such as D-Wave’s.

A central challenge in quantum computing is the variety of error and system influences that degrade system control and performance. Lacking practical and powerful enough error correction technology, the community has opted for labelling the modern class of quantum computers as noisy intermediate-scale quantum (NISQ) systems. Recognizing this is a situation likely to persist for some time, the IBM paper’s authors[I] do a nice job describing the problem and their approach to measuring performance. Excerpt:

“In these noisy intermediate-scale quantum (NISQ) systems, performance of isolated gates may not predict the behavior of the system. Methods such as randomized benchmarking, state and process tomography, and gateset tomography are valued for measuring the performance of operations on a few qubits, yet they fail to account for errors arising from interactions with spectator qubits. Given a system such as this, whose individual gate operations have been independently calibrated and verified, how do we measure the degree to which the system performs as a general purpose quantum computer? We address this question by introducing a single-number metric, the quantum volume, together with a concrete protocol for measuring it on near-term systems. Similar to how LINPACK is used for comparing diverse classical computers, this metric is not tailored to any particular system, requiring only the ability to implement a universal set of quantum gates.

“The quantum volume protocol we present is strongly linked to gate error rates, and is influenced by underlying qubit connectivity and gate parallelism. It can thus be improved by moving toward the limit in which large numbers of well-controlled, highly coherent, connected, and generically programmable qubits are manipulated within a state-of-the-art circuit rewriting toolchain. High-fidelity state preparation and readout are also necessary. In this work, we evaluate the quantum volume of current IBM Q devices, and corroborate the results with simulations of the same circuits under a depolarizing error model.”

In practice, explained Sheldon, “We generate model circuits which have a specific form where they are sequences of different layers of random entangling gates. The first step is entangling gates between different pairs of qubits on the device. Then we permute the pairing of qubits, into another layer of entangling gates. Each of these layers we call the depth. So if we have three layers, it’s depth3. What we are looking at are circuits we call square circuits with the same number of qubits as the depth in the circuit. Since we are still talking about small enough numbers of qubits that we can simulate these circuits [on classical systems].

“We run an ideal simulation of the circuit and from get a probability distribution of all the possible outcomes. At the end of applying the circuit, the system should be in some state and if we were to measure it we would get a bunch of bit streams, outcomes, with some probabilities. Then we can compare the probabilities from the ideal case to what we actually measured. Based on how close we are to the ideal situation, we say whether or not we were successful. There are details in the paper about how we actually define the success and how we compare the experimental circuits to the ideal circuits. The main point is by doing these model circuits we’re sort of representing a generic quantum algorithm – [we realize] a quantum algorithm doesn’t use random circuits but this is kind of a proxy for that,” she said.

Shown below are some data characterizing IBM systems – IBM Q System One, IBM Q Network systems “Tokyo” and “Poughkeepsie,” and the publicly-available IBM Q Experience system “Tenerife.” As noted in IBM’s blog the performance of a particular quantum computer can be characterized on two levels: metrics associated with the underlying qubits in the chip—what we call the “quantum device”—and overall full-system performance.

“IBM Q System One’s performance is reflected in some of the best/lowest error rates we have ever measured. The average two qubit gate error is less than two percent, and the best gate has less than one percent error rate. Our devices are close to being fundamentally limited by coherence times, which for IBM Q System One averages 73μs,” write Jay Gambetta (IBM Fellow) and Sheldon in the blog. “The mean two-qubit error rate is within a factor of two (x1.68) of the coherence limit, the theoretical limit set by the qubit T1 and T2 (74μs and 69μs on average for IBM Q System One). This indicates that the errors induced by our controls are quite small, and we are achieving close to the best possible qubit fidelities on this device.”

It will be interesting to see how the quantum computing community responds to the CV metric. Back in May when Hyperion Research launched its quantum practice, analyst Bob Sorensen said, “One of the things I’m hoping we can at least play a role in is the idea of thinking about quantum computing benchmarks. Right now, if you read the popular press, and I say ‘IBM’ and the first thing you think of is, yes they have a 50-qubit system. That doesn’t mean much to anybody other than it’s one more qubit than a 49-qubit system. What I am thinking about is asking these people how can we start to characterize across a number of different abstractions and implementations to gain a sense of how we can measure progress.”

IBM has high hopes for Quantum Volume.

Link to release: https://newsroom.ibm.com/2019-03-04-IBM-Achieves-Highest-Quantum-Volume-to-Date-Establishes-Roadmap-for-Reaching-Quantum-Advantage

Link to blog: https://www.ibm.com/blogs/research/2019/03/power-quantum-device/

Link to paper: https://arxiv.org/pdf/1811.12926.pdf

Feature image; IBM Q System One

[i]Validating quantum computers using randomized model circuits, Andrew W. Cross, Lev S. Bishop, Sarah Sheldon, Paul D. Nation, and Jay M. Gambetta IBM T. J. Watson Research Center, https://arxiv.org/pdf/1811.12926.pdf

 

 

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!

RIKEN Post-K Supercomputer Named After Japan’s Tallest Peak

May 23, 2019

May 23 -- RIKEN President Hiroshi Matsumoto announced that the successor to the K computer will be named Fugaku, another name for Mount Fuji, which is the tallest mountain peak in Japan. Supercomputer Fugaku, developed b Read more…

By Tiffany Trader

Cray’s Emerging Market & Technology Director Arti Garg Peers Around HPC/AI Corner

May 23, 2019

In her position as emerging market and technology director at Cray, Arti Garg doesn't just have a front-row seat to the future of computing, she plays an active role in making that future happen. Key to Garg's role is understanding how deep learning scientists are using state-of-the-art HPC infrastructures and figuring out how to push those limits further. Read more…

By Tiffany Trader

Combining Machine Learning and Supercomputing to Ferret out Phishing Attacks

May 23, 2019

The relentless ingenuity that drives cyber hacking is a global engine that knows no rest. Anyone with a laptop and run-of-the-mill computer smarts can buy or rent a phishing kit and start attacking – or it can be done Read more…

By Doug Black

HPE Extreme Performance Solutions

HPE and Intel® Omni-Path Architecture: How to Power a Cloud

Learn how HPE and Intel® Omni-Path Architecture provide critical infrastructure for leading Nordic HPC provider’s HPCFLOW cloud service.

For decades, HPE has been at the forefront of high-performance computing, and we’ve powered some of the fastest and most robust supercomputers in the world. Read more…

IBM Accelerated Insights

Who’s Driving Your Car?

Delivering a fully autonomous driving (AD) vehicle remains a key priority for both manufacturers and technology firms (“firms”). However, passenger safety is now a top-of-mind concern due in great part, to fatalities resulting from driving tests over the past years. Read more…

TACC’s Upgraded Ranch Data Storage System Debuts New Features, Exabyte Potential

May 22, 2019

There's a joke attributed to comedian Steven Wright that goes, "You can't have everything. Where would you put it?" Users of advanced computing can likely relate to this. The exponential growth of data poses a steep challenge to efforts for its reliable storage. For over 12 years, the Ranch system at the Texas Advanced Computing Center... Read more…

By Jorge Salazar, TACC

Cray’s Emerging Market & Technology Director Arti Garg Peers Around HPC/AI Corner

May 23, 2019

In her position as emerging market and technology director at Cray, Arti Garg doesn't just have a front-row seat to the future of computing, she plays an active role in making that future happen. Key to Garg's role is understanding how deep learning scientists are using state-of-the-art HPC infrastructures and figuring out how to push those limits further. Read more…

By Tiffany Trader

Combining Machine Learning and Supercomputing to Ferret out Phishing Attacks

May 23, 2019

The relentless ingenuity that drives cyber hacking is a global engine that knows no rest. Anyone with a laptop and run-of-the-mill computer smarts can buy or re Read more…

By Doug Black

Cray – and the Cray Brand – to Be Positioned at Tip of HPE’s HPC Spear

May 22, 2019

More so than with most acquisitions of this kind, HPE’s purchase of Cray for $1.3 billion, announced last week, seems to have elements of that overused, often Read more…

By Doug Black and Tiffany Trader

HPE to Acquire Cray for $1.3B

May 17, 2019

Venerable supercomputer pioneer Cray Inc. will be acquired by Hewlett Packard Enterprise for $1.3 billion under a definitive agreement announced this morning. T Read more…

By Doug Black & Tiffany Trader

Deep Learning Competitors Stalk Nvidia

May 14, 2019

There is no shortage of processing architectures emerging to accelerate deep learning workloads, with two more options emerging this week to challenge GPU leader Nvidia. First, Intel researchers claimed a new deep learning record for image classification on the ResNet-50 convolutional neural network. Separately, Israeli AI chip startup Hailo.ai... Read more…

By George Leopold

CCC Offers Draft 20-Year AI Roadmap; Seeks Comments

May 14, 2019

Artificial Intelligence in all its guises has captured much of the conversation in HPC and general computing today. The White House, DARPA, IARPA, and Departmen Read more…

By John Russell

Cascade Lake Shows Up to 84 Percent Gen-on-Gen Advantage on STAC Benchmarking

May 13, 2019

The Securities Technology Analysis Center (STAC) issued a report Friday comparing the performance of Intel's Cascade Lake processors with previous-gen Skylake u Read more…

By Tiffany Trader

Nvidia Claims 6000x Speed-Up for Stock Trading Backtest Benchmark

May 13, 2019

A stock trading backtesting algorithm used by hedge funds to simulate trading variants has received a massive, GPU-based performance boost, according to Nvidia, Read more…

By Doug Black

Cray, AMD to Extend DOE’s Exascale Frontier

May 7, 2019

Cray and AMD are coming back to Oak Ridge National Laboratory to partner on the world’s largest and most expensive supercomputer. The Department of Energy’s Read more…

By Tiffany Trader

Graphene Surprises Again, This Time for Quantum Computing

May 8, 2019

Graphene is fascinating stuff with promise for use in a seeming endless number of applications. This month researchers from the University of Vienna and Institu Read more…

By John Russell

Why Nvidia Bought Mellanox: ‘Future Datacenters Will Be…Like High Performance Computers’

March 14, 2019

“Future datacenters of all kinds will be built like high performance computers,” said Nvidia CEO Jensen Huang during a phone briefing on Monday after Nvidia revealed scooping up the high performance networking company Mellanox for $6.9 billion. Read more…

By Tiffany Trader

It’s Official: Aurora on Track to Be First US Exascale Computer in 2021

March 18, 2019

The U.S. Department of Energy along with Intel and Cray confirmed today that an Intel/Cray supercomputer, "Aurora," capable of sustained performance of one exaf Read more…

By Tiffany Trader

ClusterVision in Bankruptcy, Fate Uncertain

February 13, 2019

ClusterVision, European HPC specialists that have built and installed over 20 Top500-ranked systems in their nearly 17-year history, appear to be in the midst o Read more…

By Tiffany Trader

Intel Reportedly in $6B Bid for Mellanox

January 30, 2019

The latest rumors and reports around an acquisition of Mellanox focus on Intel, which has reportedly offered a $6 billion bid for the high performance interconn Read more…

By Doug Black

Looking for Light Reading? NSF-backed ‘Comic Books’ Tackle Quantum Computing

January 28, 2019

Still baffled by quantum computing? How about turning to comic books (graphic novels for the well-read among you) for some clarity and a little humor on QC. The Read more…

By John Russell

Deep Learning Competitors Stalk Nvidia

May 14, 2019

There is no shortage of processing architectures emerging to accelerate deep learning workloads, with two more options emerging this week to challenge GPU leader Nvidia. First, Intel researchers claimed a new deep learning record for image classification on the ResNet-50 convolutional neural network. Separately, Israeli AI chip startup Hailo.ai... Read more…

By George Leopold

Leading Solution Providers

SC 18 Virtual Booth Video Tour

Advania @ SC18 AMD @ SC18
ASRock Rack @ SC18
DDN Storage @ SC18
HPE @ SC18
IBM @ SC18
Lenovo @ SC18 Mellanox Technologies @ SC18
NVIDIA @ SC18
One Stop Systems @ SC18
Oracle @ SC18 Panasas @ SC18
Supermicro @ SC18 SUSE @ SC18 TYAN @ SC18
Verne Global @ SC18

The Case Against ‘The Case Against Quantum Computing’

January 9, 2019

It’s not easy to be a physicist. Richard Feynman (basically the Jimi Hendrix of physicists) once said: “The first principle is that you must not fool yourse Read more…

By Ben Criger

Deep500: ETH Researchers Introduce New Deep Learning Benchmark for HPC

February 5, 2019

ETH researchers have developed a new deep learning benchmarking environment – Deep500 – they say is “the first distributed and reproducible benchmarking s Read more…

By John Russell

IBM Bets $2B Seeking 1000X AI Hardware Performance Boost

February 7, 2019

For now, AI systems are mostly machine learning-based and “narrow” – powerful as they are by today's standards, they're limited to performing a few, narro Read more…

By Doug Black

Arm Unveils Neoverse N1 Platform with up to 128-Cores

February 20, 2019

Following on its Neoverse roadmap announcement last October, Arm today revealed its next-gen Neoverse microarchitecture with compute and throughput-optimized si Read more…

By Tiffany Trader

Intel Launches Cascade Lake Xeons with Up to 56 Cores

April 2, 2019

At Intel's Data-Centric Innovation Day in San Francisco (April 2), the company unveiled its second-generation Xeon Scalable (Cascade Lake) family and debuted it Read more…

By Tiffany Trader

Announcing four new HPC capabilities in Google Cloud Platform

April 15, 2019

When you’re running compute-bound or memory-bound applications for high performance computing or large, data-dependent machine learning training workloads on Read more…

By Wyatt Gorman, HPC Specialist, Google Cloud; Brad Calder, VP of Engineering, Google Cloud; Bart Sano, VP of Platforms, Google Cloud

In Wake of Nvidia-Mellanox: Xilinx to Acquire Solarflare

April 25, 2019

With echoes of Nvidia’s recent acquisition of Mellanox, FPGA maker Xilinx has announced a definitive agreement to acquire Solarflare Communications, provider Read more…

By Doug Black

Nvidia Claims 6000x Speed-Up for Stock Trading Backtest Benchmark

May 13, 2019

A stock trading backtesting algorithm used by hedge funds to simulate trading variants has received a massive, GPU-based performance boost, according to Nvidia, Read more…

By Doug Black

  • arrow
  • Click Here for More Headlines
  • arrow
Do NOT follow this link or you will be banned from the site!
Share This