D-Wave Reports Major Performance Gains for Stride Solver

D-Wave Quantum has enhanced the core algorithm of its Stride hybrid solver, reporting improved run times and solution quality across various optimization benchmarks. The company, which offers both annealing and gate-model systems, released these findings on November 4. These advancements are particularly significant as they address large routing and scheduling problems where traditional methods often falter. Additionally, D-Wave has launched new training initiatives to encourage developer adoption.

Benchmark Results

The updated Stride solver operates within D-Wave’s Leap quantum cloud service, integrating quantum and classical computing resources to tackle a single problem. D-Wave states that the enhanced version achieved superior solutions in 15 seconds, a task that previously took 120 seconds. All subsequent figures are based on D-Wave’s internal benchmark testing.

• Traveling Salesperson Problems (up to 10,000 locations): Solutions averaged within 0.02% of the optimal result in 15 seconds, demonstrating a more than 200-fold improvement in proximity to optimal compared to the previous solver at the same runtime.
• Traveling Salesperson Problems (over 10,000 locations): Solutions averaged within 3.8% of optimal in 120 seconds, representing a more than 500-fold improvement in proximity to optimal compared to the previous solver at equal runtime.
• Assembly Line Balancing: For straightforward cases, the enhanced solver surpassed the previous version’s 120-second output in just 15 seconds.
• Applied Quantum Optimization Learning Path: A new three-course learning path, now available through Leap, covers use-case identification, mathematical modeling, and application development.

It’s important to note that these comparisons are internal, evaluating the new solver against D-Wave’s prior version, rather than against competing classical optimizers. Therefore, these results should be interpreted as version-over-version improvements, not a direct measure of quantum advantage.

Driving Adoption

D-Wave is positioning Stride for applications in logistics, manufacturing, and resource allocation, especially where plans are dynamic and large-scale instances are common. Alex Condello, Senior Vice President of Algorithms, Performance, and Tools at D-Wave, stated in the release, “Developers need proven technology that finds better answers, faster.” The new courses, which include Leap access and office hours for certain tiers, indicate the company’s focus on developer onboarding as a key factor for broader adoption.

This update follows recent commercial developments, including a partnership with LG CNS for quantum optimization in South Korea. For more information on hybrid optimization and its commercial applications, visit QNews at qnews.io.

Read more

• Infleqtion Field Tests Optical Atomic Clock With US Navy
• IQM Reports Full-Chip Entanglement Across 54 Qubits