D-Wave Systems has shown its new 5000 qubit quantum annealing system that uses a vastly improved Pegasus architecture.
Nextbigfuture wrote about the Pegasus architecture and the 5,640 qubit D-Wave prototype back in mid-January. D-Wave has now officially announced the 5000+ qubit system and the new architecture.
The next-generation D-Wave quantum platform will include ongoing updates available through the cloud, a new quantum system, and new software for tools and applications:
New Topology: The Pegasus™ topology is the most connected of any commercial quantum system in the world. Currently, each qubit in the Chimera™ topology is connected to six other qubits. With the Pegasus topology, each qubit is connected to 15 other qubits. With two and a half times more connectivity, Pegasus enables the embedding of larger problems with fewer physical qubits than the Chimera topology.
The D-Wave Ocean™ software development kit (SDK) includes tools for generating the Pegasus topology. Interested users can try embedding their problems on Pegasus.
Solve Larger Problems More Efficiently with Pegasus
Here is a 24-page technical white paper on the Pegasus Architecture.
Heuristic embeddings. D-Wave comparison of embeddings in P6 (with 680 qubits and 4484 couplers) versus C16 (with 2048 qubits and 6016 couplers) for a diverse set of problems shows Pegasus consistently achieving around a 50-60% reduction in chain length over Chimera.
Native structured Ising models. Simple Pegasus benchmarks differ qualitatively from the corresponding Chimera ones, indicating that the topology may support larger energy barriers and thereby offer more opportunities for differentiation between classical and quantum dynamics. Techniques commonly in use to accelerate optimization (and sampling) in Chimera-structured problems seem less potent on Pegasus structured problems.
More Qubits and Other Improvements
Lower Noise: The next-generation system will include the lowest-noise commercially-available quantum processing units (QPUs) ever produced by D-Wave. This new QPU technology further improves qubit coherence, and paves the way to even greater speedups for existing and new quantum computing applications.
Increased Qubit Count: With more than 5000 qubits, the next-generation platform will more than double the qubit count of the existing D-Wave 2000Q™ system. Combining more qubits with the new topology will give quantum programmers access to a larger, denser, and more powerful graph for building commercial quantum applications.
Expansion of Hybrid Software & Tools: Further investments in ease-of-use, automation, and new tools will provide an even more powerful hybrid rapid development environment. The hybrid software and tools will build on the existing D-Wave Hybrid™ offering, allowing developers to run across both classical and the next-generation quantum platforms in the familiar coding environment, Python. The modular approach incorporates logic to simplify distribution of classical and quantum tasks, allowing developers to interrupt processing and synchronize across systems in order to draw maximum computing power out of each system. Additionally, the open source tools in D-Wave’s Ocean SDK are written in Python and C for ease of use and performance.
Flexible Access: The next-generation D-Wave quantum platform will be available to run and build applications by purchasing hours of use through the Leap™ cloud service, and the next-generation system will also be available for installation at customer sites. Flexible purchase plans allow developers, researchers, governments, institutions, and forward-thinking businesses to get started and access the D-Wave quantum system in the way that works for them and their business.
Ongoing Releases: Components of the D-Wave next-generation quantum platform will come to market between now and mid-2020 via ongoing QPU and software updates to be made available through the cloud. The complete system will be available through cloud access and for on-premise installation in mid-2020. Interested users and customers can get started today by using the Ocean tools to embed problems on the new Pegasus topology and will benefit from new components of the platform through Leap as they become available at https://cloud.dwavesys.com/leap/login/.
D-Wave’s next-generation quantum computing platform has hardware, software, and tools to accelerate and ease the delivery of quantum computing applications. Reflecting years of customer feedback, the platform captures users’ priorities and business requirements and will deliver significant performance gains and greater solution precision. Users will also benefit from the ability to embed larger problems on the quantum computer, speeding the development of commercial quantum applications.
D-Wave’s next-generation platform will be a significant milestone in D-Wave’s commitment to help customers succeed in their real-world quantum computing application development efforts. To date, D-Wave customers have developed more than 100 early applications in areas as diverse as airline scheduling, election modeling, quantum chemistry simulation, automotive design, preventative healthcare, logistics, and more. Many have also developed software tools that simplify application development. These existing applications, tools, and community give developers a wealth of examples to learn from and build upon. The next-generation platform promises to continue to expand the variety, performance, precision, and breadth of quantum applications.
Comparison of 2000 Qubit D-Wave Annealing to 19 Qubit Rigetti
D-Wave’s 2000 qubit system vastly outperformed the 19 qubit rigetti quantum computer.
The general expectation is that a 2000 qubit D-Wave system would match the performance of a 44 noisy approximate superconducting systems like the Rigetti, Google or Intel. It is about the square root of the D-Wave qubits. D-Wave will have its 5000 to 5600 Qubit system commercially available in mid-2020. D-Wave should match up well against 70-80 qubit systems from the competing systems with cleaner qubits.
SOURCES- D-Wave Systems
Written By Brian Wang. Nextbigfuture.com
Brian Wang is a Futurist Thought Leader and a popular Science blogger with 1 million readers per month. His blog Nextbigfuture.com is ranked #1 Science News Blog. It covers many disruptive technology and trends including Space, Robotics, Artificial Intelligence, Medicine, Anti-aging Biotechnology, and Nanotechnology.
Known for identifying cutting edge technologies, he is currently a Co-Founder of a startup and fundraiser for high potential early-stage companies. He is the Head of Research for Allocations for deep technology investments and an Angel Investor at Space Angels.
A frequent speaker at corporations, he has been a TEDx speaker, a Singularity University speaker and guest at numerous interviews for radio and podcasts. He is open to public speaking and advising engagements.
I agree. But as they add more qubits they will run in to the same problems everyone else runs in to.
So 5600 qubits is effectively 75 qubits. Adding more Dwave qubits is not going to improve performance significantly. Performance improvement is going to come from connecting more of the qubits they already have to each other.
Wow… take about QPU envy… my GPU needs a QPU!