QGraph Solves the Most Challenging Graph Problems on Quantum Computers without Complex Quantum Programming
Quantum Computing Inc., a leader in bridging the power of classical and quantum computing, has launched QGraph which analyzes graphs (i.e., collections of vertices and edges) with QCI’s cloud-based Qatalyst ready-to-run quantum software. QGraph together with Qatalyst enables users and analysts to solve the most computationally expensive graph problems – the kind that can benefit the most from quantum computing, that are essential for understanding high-dimensional data in fast moving contexts, and that to date have been prohibitively expensive to compute in practice – quickly and cost-effectively on both classical (CPU) and quantum (QPU) computers.
Recommended ITech News: AirTrunk Opens Second Sydney Hyperscale Data Centre, Cementing Leadership in the Australian Market
As part of Qatalyst, QGraph performs the most challenging graph kernels via the powerful Q API, transforms input graphs into constrained optimization problems, and then delivers them to the Qatalyst Core computational engine for CPU and/or QPU processing. The process employs quantum-ready techniques that fuel increases in accuracy and deliver a diversity of valuable results.
Graphs offer a powerful way to analyze heterogeneous data that has many dimensions, is unstructured and has sparse values across all variables. Ridesharing and food delivery services are two perfect examples where graphs can be highly useful. They both calculate physical location data to efficiently match customers with providers, and problems of these types are more easily solved visually with a graph. Graphs are also often used to model networks, such as social, metabolic, gene and transportation, as well as molecular structures.
Other business use cases of graph analytics include air traffic control and route optimization for efficiency and lowering fuel consumption. Retailers use graph analytics to determine what products are frequently purchased together and by what type of customer, enabling better marketing and sales intelligence. Healthcare and pharmaceutical companies use graph analytics when examining patient symptoms and outcomes for medical analysis and drug development.
Some of the most valuable calculations are highly compute-intensive and grow exponentially with the size of the graph. The complexity of data and relationships within large computations can become so great that classical systems have difficulty solving them. This limitation is the reason many subject matter experts (SMEs) are unable to perform these analyses on their graphs despite the benefits.
Recommended ITech News: Intermedia Announces Launch of Initial Public Offering
“QGraph takes graph analytics to the next level, leveraging the power of quantum-ready constrained optimization to easily and more cost-effectively solve the most intractable problems,” stated Robert Liscouski, CEO of QCI. “SMEs can continue to use well-known graph functions and constructs without any new programming, low-level coding, or changes to their models. We believe this capability makes our ready-to-run quantum software exceptionally valuable and unique.”
QGraph enables SMEs and programmers to work with familiar graphs and functions, including graph partitioning, minimum clique cover, and community detection. After the graph is submitted to the Qatalyst API, which implements familiar NetworkX-type functions, QGraph automatically transforms the graph into a constrained optimization problem based on the specific requested function. The problem is then submitted to the Qatalyst Core for quantum transformation and processing. When results are returned, QGraph transforms and presents them back to the requesting workflow, application, or SME in a graph-relevant format.
“Qatalyst leverages the power of partitioning as part of its iterative, mathematical optimization of results within its quantum optimization engine,” noted Steve Reinhardt, vice president of product development at QCI. “We believe Qatalyst’s ability to partition very large graphs, iteratively process these partitions in dynamic mathematical ways, and then recombine for highest quality results, is unparalleled in the quantum software world. With QGraph, users now have a practical way to solve some of the most computationally advanced and expensive graph problems.”
Recommended ITech News: Alluxio Advances Analytics and AI with NVIDIA Accelerated Computing
