D-Wave to Launch Advantage Quantum Computing Platform

D-Wave Systems Inc. has announced that its next-generation quantum computing system, which has been named Advantage, will be available in the company’s Leap quantum cloud access environment in mid-2020. Advantage, which is D-Wave’s fifth system upgrade, will incorporate hardware, software, and tools which should accelerate and ease the delivery of quantum computing applications, especially for larger problems less accessible by conventional systems.
D-Wave has stated that the platform incorporates user feedback on priorities and business requirements, and should deliver significant performance gains and greater solution precision. The new generation of hardware also lowers latency, which reduces the time it takes for users to receive results from D-Wave’s hardware.

Pegasus Topology Replaces Chimera

The upgrades to be offered by Advantage include a new topology, Pegasus, which will replace the currently used Chimera topology. Each qubit in Chimera is connected to six other qubits, while in Pegasus, each qubit will be connected to 15 other qubits. This 250% more connectivity will allow the embedding of larger problems with fewer physical qubits than the Chimera topology.

Advantage will use 5000 qubits, more than double the 2048 qubit count of the existing D-Wave 2000Q system. The combination of more qubits with the Pegasus topology will give programmers access to a larger, denser, and more powerful graph for building commercial quantum applications.

System Includes Reduced Qubit Noise

The current 2000Q computer has about 100,000 Josephson junctions — Advantage will have over one million. In addition, it will have more than 100 meters of wiring and the chip will be four times larger than the 2000Q. The current generation of chips has 6,000 connections among its 2,000 qubits, while the next system will have 40,000 connections for its 5,000 qubits.

Advantage will also include the lowest-noise commercially available quantum processing units (QPUs) ever produced by D-Wave. The noise of individual qubits will be lowered by about three- to four-fold, which will have significant impact on tunneling. Lowered noise also yields improved performance, precision, and impact on quantum effects.

This new technology was already introduced last May to Leap cloud users in the D-Wave 2000Q system. The company has published research demonstrating a 25 times speed-up of the new processor as compared to its predecessor, which involved posing a common set of identical spin glass problems to two such QPUs. The experimental results confirmed a positive correlation between reduced noise and improved performance with the obsevation of at least a 25 times speed-up in solving spin glass problems.

Hybrid Software and Tools Built into Platform

Advantage will also provide an expansion of hybrid software and tools, which will be built on the existing D-Wave Hybrid open source workflow platform. This expansion was released last June and allows developers to rapidly develop hybrid applications that can run across classical and D-Wave 2000Q quantum systems in the Python coding environment.

A modular approach incorporates logic to simplify distribution of classical and quantum tasks, allowing developers to interrupt and synchronize across the systems and draw maximum computing power out of each system. Advantage also is capable of breaking down large problems that are bigger than the QPU into piece parts that are then recombined for the overall solution.

LANL Signed On as First Customer

The new platform, which will be available through sale or lease as a standalone system, or via the Leap quantum cloud service, has already won its first customers. The Los Alamos National Lab (LANL) has signed a contract to upgrade its standalone system to Advantage upon its release. D-Wave also announced its first European Leap quantum cloud site at the Jülich Supercomputing Center in Germany, offering practical quantum computers to government and academic customers in Germany and throughout Europe.

D-Wave’s quantum system is different from those being developed by other companies in that it runs a quantum annealing algorithm to find the lowest points in a virtual energy landscape representing a computational problem to be solved. The lowest points in the landscape correspond to optimal or near-optimal solutions to the problem.

The extent to which D-Wave’s computing technology qualifies as quantum technology has generated debate by some academics. In particular, some researchers question whether D-Wave’s systems have shown quantum entanglement, an indicator that it relies on quantum effects.