Fast Circuit Simulation Based on Parallel-Distributed LIM using Cloud Computing System

Abstract

Abstract-This paper describes a fast circuit simulation technique using the latency insertion method (LIM) with a parallel and distributed leapfrog algorithm. The numerical simulation results on the PC cluster system that uses the cloud computing system are shown. As a result, it is confirmed that our method is very useful and practical.

Locally implicit block leapfrog scheme for PDN simulation

“…Since all of the terms in the right hand sides of the updating formulas (1) and (2) can be given as the ones at the (n+1)-th or the (n+1/2)-th time step, each variable is updated explicitly, in other words, by substituting only the values at the past time steps. For these years, a variety of studies related to LIM have been done and several kinds of skillful techniques, namely, multi-rate LIM [7], cloud and GPU-LIM [11] [12], block-LIM [13], ADE-LIM [14] and parallel-distributed block-LIM [15], have been proposed, which exploit the multi-rate behavior in the time domain, the cloud/GPU computing environment, the characteristics of strongly coupled multi-conductor transmission lines, alternating-direction explicit scheme and the parallelism of the block-LIM, respectively.…”

“…In order to solve (11), the direct method such as LU decomposition is required. However, similarly to the block LIM [13], because the direct method is applied to only the local (smaller) block submatrices, the total amount of computational cost is much smaller than the conventional method such as SPICE.…”

Proposal of an Effective Computation Environment for the Traveling Salesman Problem Using Cloud Computing

“…Google Maps is a tool used worldwide that allows users to extract all sorts of data. We are able to visualize our results through Google Maps for (iii) and using a cloud environment will allow us to handle any size calculation for (iv) (1) .…”

“…Because the work is spread amongst several computers, it provides a large-scale environment for solving combination optimized calculations without a high cost (1) . In our view, the primary benefit of this is being able to use of constantly updated map information.…”