Abstract-The conjugate gradient (CG) is one of the most widely used iterative methods for solving systems of linear equations. However, parallelizing CG for large sparse systems is difficult due to the inherent irregularity in memory access pattern. We propose a novel processor architecture for the sparse conjugate gradient method. The architecture consists of multiple processing elements and memory banks, and is able to compute efficiently both sparse matrix-vector multiplication, and other dense vector operations. A Beneš permutation network with an optimised control scheme is introduced to reduce memory bank conflicts without expensive logic. We describe a heuristics for offline scheduling, the effect of which is captured in a parametric model for estimating the performance of designs generated from our approach.
In the field of water distribution systems the EPANET 2 toolkit is considered nowadays the industry standard for hydraulic modelling. Unfortunately, the design and programming model of EPANET 2 have some limitations that make any attempt to extend its hydraulic solver, add new functionalities or improve performance difficult to achieve and time consuming. A new software toolkit for water distribution system modelling, CWSNet, is presented. CWSNet is developed in C++ using the object-oriented programming model. The aim is to deliver an open-source substitute for EPANET 2 that obtains numerically comparable results while providing similar or better performance, a higher degree of extensibility, as well as backward compatibility where possible. The idea behind this project is to simplify development and testing of new hydraulic elements (specific types of valves, pumps, etc) and computational algorithms (pressure-driven approaches, etc.) by keeping logically independent parts of the code separate. This also allows the performance and accuracy of new computational methods as well as the use of advanced programming techniques (multi-threading, OpenMP, GPGPU, etc) to be studied without the need for extensive code refactoring. The basic version of CWSNet gives numerically the same results as EPANET 2 for various networks while allowing the following: (a) to change the topology of the network at runtime; (b) to run different simulations of the same network or different networks in parallel (thread-safe); (c) to easily change the mathematical model and other particulars behind the hydraulic simulation engine; (d) to allow a high degree of customisation of the output of an extended period simulation. The CWSNet software capabilities are demonstrated using several examples. The results obtained demonstrate the effectiveness and efficiency of the proposed approach.
We demonstrate the use of dataflow technology in the computation of the correlation energy in molecules at the Møller-Plesset perturbation theory (MP2) level. Specifically, we benchmark density fitting (DF)-MP2 for as many as 168 atoms (in valinomycin) and show that speed-ups between 3 and 3.8 times can be achieved when compared to the MOLPRO package run on a single CPU. Acceleration is achieved by offloading the matrix multiplications steps in DF-MP2 to Dataflow Engines (DFEs). We project that the acceleration factor could be as much as 24 with the next generation of DFEs.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.