Google Charts Two-Dimensional Quantum Course

By Tiffany Trader

April 26, 2018

Quantum error correction, essential for achieving universal fault-tolerant quantum computation, is one of the main challenges of the quantum computing field and it’s top of mind for Google’s John Martinis. Delivering a presentation last week at the HPC User Forum in Tucson, Martinis, one of the world’s foremost experts in quantum computing, emphasized that building a useful quantum device is not just about the number of qubits; getting to 50 or 1,000 or 1,000,000 qubits doesn’t mean anything without quality error-corrected qubits to start with.

Martinis compares focusing on merely the number of qubits to wanting to buy a high-performance computer and only specifying the number of cores. How to create quality qubits is something that the leaders in the quantum space at this nascent stage are still figuring out. Google — as well as IBM, Intel, Rigetti, and Yale – are advancing the superconducting qubit approach to quantum computing. Microsoft, Delft, and UC Santa Barbara are involved in topological quantum computing. Photonic quantum computing and trapped ions are other approaches.

The reason quality is difficult in the first place is that qubits – the processing unit of the quantum system – are fundamentally sensitive to small errors, much more so than the classical bit. Martinis explains with a coin on the table analogy:

“If you want to think about classical bits – you can think of that as a coin on a table; we can represent classical information as heads or tails. Classical information is inherently stable. You have this coin on the table, there’s a restoring force, there’s dissipation so even if there’s a little bit of noise it’s going to be stable at zero or one. In a quantum computer you can represent [a quantum bit] not as a coin on a table but a coin in free space, where say zero is up, and one is down and rotated 90 degrees is zero plus one; and in fact you can have any amount of zero and one and it can rotate in this way to change something called quantum phase. You see since it’s kind of an analog system, it can point in any direction. This means that any small change in this is going to give you an error.

“Error correction in quantum systems is a little bit similar to what you see in classical systems where you have clocked logic so you have a memory source, where you have a clock and every clock period you can compute through some arithmetic logic and then you sequence through this and the clock timing kind of takes care of all the different delays you have in the logic. Similar here, you have kind of repetition of the error correction, based on taking the qubit and encoding it in many other qubits and doing parity measurements to see if you’re having both bit-flip errors going like this or phase flip errors going like that.”

The important thing to remember says Martinis is that if you want to have small errors, exponentially small errors, of 10-9 or 10-12, you need a lot of qubits, i.e., quantity, and pretty low error rates of about one error in one-thousand operations, i.e., quality.

In Martinis’s view, quantum computing is “a two-dimensional horse race,” where the tension between quality and quantity means you can’t think in terms of either/or; you have to think about doing both of them at the same time. Progress of the field can thus be charted on a two-dimensional plot.

The first thing to note when assessing the progress in the field are the limiting error rate and the number of qubits for a single device, says Martinis. The chart depicts, for a single device, the worst error rate, the limiting error rate, and the number of qubits. Google is aiming for an error correction of 10-3 in about 103 qubits.

“What happens, “says Martinis, “is as that error rate goes up the number of qubits you have to have to do error correction properly goes up and blows up at the error correction threshold of about 1 percent. I call this the error correction gain. It’s like building transistors with gain; if you want to make something useful you have to have an error correction that’s low enough. If the error correction is not good enough, it doesn’t matter if you have a billion qubits, you are never going to be able to make them more accurate.”

Up to 50 qubits is classically simulatable, but if the error rate is high it gets easier but it is not useful. Pointing to the lower half of the chart, Martinis says “we want to be down here and making lots of qubits. It’s only once you get down here [below the threshold] that talking quantity by itself makes sense.”

One of the challenges of staying under that error correction threshold is that scaling qubits itself can impede error correction, due to undesired cross-talk between qubits. Martinis says that the UC Santa Barbara technology it is working with was designed to reduce cross-talk to produce a scalable technology. For flux cross-talk, fledgling efforts were at 30-40 percent cross-talk. “The initial UC Santa Barbara device was between 1 percent to .1 percent cross-talk and now it’s 10-5,” says Martinis, adding “we barely can measure it.”

The solid black dot on the chart (above) represents that UC Santa Barbara chip. It is 9 qubits and dips just beneath the error correction threshold. Now with its follow-on Bristlecone chip architecture, Google is working to scale the UCSB prototype to >50 qubits to show quantum supremacy, the point at which it would be longer feasible to classically simulate it. The Google team is focused on improving error correction with the expectation that near-term applications will then be feasible. Martinis says the next step is to move out to ~1,000 qubits with exponentially small errors. The end goal is to scale up to a million-or-so qubits with low error rates to solve real-world problems that are intractable on today’s best supercomputers.

The Bristlecone chip consists of 72 qubits, arranged in 2D array. The device has been made and is now undergoing testing to make sure it is operating correctly. Google uses its Qubit Speckle algorithm to validate its quantum supremacy experiments.

Martinis reports that progress on quantum algorithms is also advancing. One of the most compelling applications for quantum computers is quantum chemistry. It’s a natural application for quantum computing, says Martinis. The algorithm though is exponentially hard. In 2011, Microsoft’s quantum computing group documented an O(n11) quantum chemistry algorithm, which would take the age of the universe to run. Work has since progressed and recently the Google theory group showed an algorithm that is Õ(N2.67) for the exact solution and O(N) for the approximate. “[The exact implementation] would take about 100 logical qubits, requiring a million physical qubits,” Martinis notes. “It’s beyond what we can do now, but now the numbers are reasonable so we can think about doing it.”

In closing, Martinis points out that the metrics for assessing the progress of quantum computing in addition to quality and quantity also include speed and connectivity. In different technologies, there can be a factor of 105 or so different speeds. For networking, he says you need at least 2D nearest neighbor corrections to do the error correction properly. Referring to the chart on Google’s key metrics (at left), Martinis says the company isn’t ready to talk about Bristlecone’s error-correction or speed yet but it anticipates good numbers and hopes to show quantum supremacy “very soon.”

Link to slides: https://www.hpcuserforum.com/presentations/tuscon2018/QCOverview_Google_UFTucson2018.pdf

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!

ARM, Fujitsu Targeting Open-source Software for Power Efficiency in 2-nm Chip

July 19, 2024

Fujitsu and ARM are relying on open-source software to bring power efficiency to an air-cooled supercomputing chip that will ship in 2027. Monaka chip, which will be made using the 2-nanometer process, is based on the Read more…

SCALEing the CUDA Castle

July 18, 2024

In a previous article, HPCwire has reported on a way in which AMD can get across the CUDA moat that protects the Nvidia CUDA castle (at least for PyTorch AI projects.). Other tools have joined the CUDA castle siege. AMD Read more…

Quantum Watchers – Terrific Interview with Caltech’s John Preskill by CERN

July 17, 2024

In case you missed it, there's a fascinating interview with John Preskill, the prominent Caltech physicist and pioneering quantum computing researcher that was recently posted by CERN’s department of experimental physi Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to communities across the globe. As climate change is warming ocea Read more…

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical claims. A paper published on July 10 by researchers in the U. Read more…

Belt-Tightening in Store for Most Federal FY25 Science Budets

July 15, 2024

If it’s summer, it’s federal budgeting time, not to mention an election year as well. There’s an excellent summary of the curent state of FY25 efforts reported in AIP’s policy FYI: Science Policy News. Belt-tight Read more…

SCALEing the CUDA Castle

July 18, 2024

In a previous article, HPCwire has reported on a way in which AMD can get across the CUDA moat that protects the Nvidia CUDA castle (at least for PyTorch AI pro Read more…

Aurora AI-Driven Atmosphere Model is 5,000x Faster Than Traditional Systems

July 16, 2024

While the onset of human-driven climate change brings with it many horrors, the increase in the frequency and strength of storms poses an enormous threat to com Read more…

Shutterstock 1886124835

Researchers Say Memory Bandwidth and NVLink Speeds in Hopper Not So Simple

July 15, 2024

Researchers measured the real-world bandwidth of Nvidia's Grace Hopper superchip, with the chip-to-chip interconnect results falling well short of theoretical c Read more…

Shutterstock 2203611339

NSF Issues Next Solicitation and More Detail on National Quantum Virtual Laboratory

July 10, 2024

After percolating for roughly a year, NSF has issued the next solicitation for the National Quantum Virtual Lab program — this one focused on design and imple Read more…

NCSA’s SEAS Team Keeps APACE of AlphaFold2

July 9, 2024

High-performance computing (HPC) can often be challenging for researchers to use because it requires expertise in working with large datasets, scaling the softw Read more…

Anders Jensen on Europe’s Plan for AI-optimized Supercomputers, Welcoming the UK, and More

July 8, 2024

The recent ISC24 conference in Hamburg showcased LUMI and other leadership-class supercomputers co-funded by the EuroHPC Joint Undertaking (JU), including three Read more…

Generative AI to Account for 1.5% of World’s Power Consumption by 2029

July 8, 2024

Generative AI will take on a larger chunk of the world's power consumption to keep up with the hefty hardware requirements to run applications. "AI chips repres Read more…

US Senators Propose $32 Billion in Annual AI Spending, but Critics Remain Unconvinced

July 5, 2024

Senate leader, Chuck Schumer, and three colleagues want the US government to spend at least $32 billion annually by 2026 for non-defense related AI systems.  T Read more…

Atos Outlines Plans to Get Acquired, and a Path Forward

May 21, 2024

Atos – via its subsidiary Eviden – is the second major supercomputer maker outside of HPE, while others have largely dropped out. The lack of integrators and Atos' financial turmoil have the HPC market worried. If Atos goes under, HPE will be the only major option for building large-scale systems. Read more…

Everyone Except Nvidia Forms Ultra Accelerator Link (UALink) Consortium

May 30, 2024

Consider the GPU. An island of SIMD greatness that makes light work of matrix math. Originally designed to rapidly paint dots on a computer monitor, it was then Read more…

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_1687123447

Nvidia Economics: Make $5-$7 for Every $1 Spent on GPUs

June 30, 2024

Nvidia is saying that companies could make $5 to $7 for every $1 invested in GPUs over a four-year period. Customers are investing billions in new Nvidia hardwa Read more…

Nvidia Shipped 3.76 Million Data-center GPUs in 2023, According to Study

June 10, 2024

Nvidia had an explosive 2023 in data-center GPU shipments, which totaled roughly 3.76 million units, according to a study conducted by semiconductor analyst fir Read more…

AMD Clears Up Messy GPU Roadmap, Upgrades Chips Annually

June 3, 2024

In the world of AI, there's a desperate search for an alternative to Nvidia's GPUs, and AMD is stepping up to the plate. AMD detailed its updated GPU roadmap, w Read more…

Some Reasons Why Aurora Didn’t Take First Place in the Top500 List

May 15, 2024

The makers of the Aurora supercomputer, which is housed at the Argonne National Laboratory, gave some reasons why the system didn't make the top spot on the Top Read more…

Intel’s Next-gen Falcon Shores Coming Out in Late 2025 

April 30, 2024

It's a long wait for customers hanging on for Intel's next-generation GPU, Falcon Shores, which will be released in late 2025.  "Then we have a rich, a very Read more…

Leading Solution Providers

Contributors

Google Announces Sixth-generation AI Chip, a TPU Called Trillium

May 17, 2024

On Tuesday May 14th, Google announced its sixth-generation TPU (tensor processing unit) called Trillium.  The chip, essentially a TPU v6, is the company's l 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…

IonQ Plots Path to Commercial (Quantum) Advantage

July 2, 2024

IonQ, the trapped ion quantum computing specialist, delivered a progress report last week firming up 2024/25 product goals and reviewing its technology roadmap. 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…

The NASA Black Hole Plunge

May 7, 2024

We have all thought about it. No one has done it, but now, thanks to HPC, we see what it looks like. Hold on to your feet because NASA has released videos of wh 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…

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…

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…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire