Join us (virtually) from May 24 – 28 at the 12th International Particle Accelerator Conference (IPAC’21) where Reservoir Labs’ Dr. Harper Langston, Technical Lead, Algorithms & Sensors, presents the poster titled “MACH-B: Fast Multipole Method Approaches in Particle Accelerator Simulations
On March 26 at 9:00pm CET / 4:00pm ET, Reservoir Labs’ Dr. Benoit Meister presents recent research on “Trading off Latency for Memory Bandwidth in Spatial Architectures using the Polyhedral Model” at the International Conference on HPC & Simulation (HPCS),
On January 20, 2021 at the 11th International Workshop on Polyhedral Compilation Techniques (IMPACT), Reservoir Labs presents its latest paper, “Static Versioning in the Polyhedral Model,” authored by Adithya Dattatri and Dr. Benoit Meister. This workshop is in conjunction
Advanced computing hardware and systems require advanced compilers. Learn more about our boutique solutions powered by decades of innovation.
Providing solutions based on rock solid, enterprise-class, high speed network sensing and spectral hypergraph analytics - using advanced algorithms and mathematics to give experts an advantage.
Based on breakthrough research, explore our suite of technologies addressing network flow optimization and packet path acceleration.
Cutting-edge hypergraph analysis technology for Big Data spanning security, finance, geospatial applications, and biology.
Advanced mathematics for new algorithms with asymptotic advantages, for problems in data analytics, sensing, and cognitive systems.
The MACH-B (Multipole Accelerator Codes for Hadron Beams) project is developing a Fast Multipole Method [1–7] (FMM)-based tool for higher fidelity modeling of particle accelerators for high-energy physics within the next generation of Fermilab’s Synergia simulation package [8]. MACH-B incorporates (1) highly-scalable, high-performance and generally-applicable FMM-based algorithms [5–7, 9] to
We present a fast algorithm for transforming loop nests represented in the polyhedral model, which have been placed onto a spatial grid of processors. The goal of the transformation is to avoid broadcasts, which can put extra pressure on remote memory bandwidth, and turn them mainly into neighbor-to-neighbor reuse. The
In a system for storing in memory a tensor that includes at least three modes, elements of the tensor are stored in a mode-based order for improving locality of references when the elements are accessed during an operation on the tensor. To facilitate efficient data reuse in a tensor transform
632 Broadway, Suite 803
New York, New York 10012
