Integrated circuit simulation multi-thread management parallel method and device with secrecy function

Abstract

The invention provides an integrated circuit simulation multi-thread management parallel method and device with a secrecy function. The method comprises the steps that electromagnetic simulation of a whole integrated circuit cloud platform is divided into two parts, namely a cloud computing platform and a client; the client extracts integrated circuit model information and calculation conditions which need to be calculated, and parallel coarse particles are formed and sent to the cloud platform; the cloud platform builds a management process, the management process builds a mutual exclusion body and a thread, and the thread builds a calculation process according to the state of the working mutual exclusion body and distributes and manages calculation coarse particles; and the threads created by the calculation process and the management process realize query and response communication to complete the distributed calculation coarse particles. According to the invention, only the parallel coarse particles extracted by the client are sent to the cloud platform instead of sending all information of the whole integrated circuit model to the cloud platform, so that the integrated circuit model is prevented from being leaked out through the Internet; and the calculation coarse particles are distributed and managed by adopting a mutual exclusion body and a thread, so that the parallel calculation efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of integrated circuit simulation, in particular to a parallel method and device for integrated circuit simulation multi-thread management with a security function. Background technique [0002] The ultra-large-scale parallel computing based on the cloud computing platform is the trend of multi-user, multi-task, and multi-model large-scale simulation in the future. Under the cloud computing platform, different users are required to submit multiple computing tasks of different models through the network at different times. These computing tasks form a computing queue, and are finally distributed in order to run on clusters with thousands of cores or more than 10,000 cores. How to reduce Computing the waiting time of nodes and efficiently running computing tasks submitted by users, thus reducing the waiting time of computing tasks in the queue are not only the most concerned issues for users who submit computin...

AI Technical Summary

Problems solved by technology

However, building a parallel computing environment with strong computing power requires a large construction cost and requires dedicated personnel to maintain it. Therefore, it is urgent to design a design that can not only ensure that the designed integrated circuit model is not leaked, but also can be implemented at low cost and quickly through cloud computing. simulation method

Images