SimNet: Computer Architecture Simulation using Machine Learning

05/12/2021
by   Lingda Li, et al.
6

While cycle-accurate simulators are essential tools for architecture research, design, and development, their practicality is limited by an extremely long time-to-solution for realistic problems under investigation. This work describes a concerted effort, where machine learning (ML) is used to accelerate discrete-event simulation. First, an ML-based instruction latency prediction framework that accounts for both static instruction/architecture properties and dynamic execution context is constructed. Then, a GPU-accelerated parallel simulator is implemented based on the proposed instruction latency predictor, and its simulation accuracy and throughput are validated and evaluated against a state-of-the-art simulator. Leveraging modern GPUs, the ML-based simulator outperforms traditional simulators significantly.

READ FULL TEXT
research
05/02/2020

Minerva: A Portable Machine Learning Microservice Framework for Traditional Enterprise SaaS Applications

In traditional SaaS enterprise applications, microservices are an essent...
research
10/15/2018

MGSim + MGMark: A Framework for Multi-GPU System Research

The rapidly growing popularity and scale of data-parallel workloads dema...
research
04/18/2021

CodeAPeel: An Integrated and Layered Learning Technology For Computer Architecture Courses

This paper introduces a versatile, multi-layered technology to help supp...
research
04/03/2023

X-TIME: An in-memory engine for accelerating machine learning on tabular data with CAMs

Structured, or tabular, data is the most common format in data science. ...
research
01/15/2022

Calipers: A Criticality-aware Framework for Modeling Processor Performance

Computer architecture design space is vast and complex. Tools are needed...
research
06/14/2023

Transpiling RTL Pseudo-code of the POWER Instruction Set Architecture to C for Real-time Performance Analysis on Cavatools Simulator

This paper presents a transpiler framework for converting RTL pseudo cod...
research
09/17/2023

Fully Convolutional Generative Machine Learning Method for Accelerating Non-Equilibrium Greens Function Simulations

This work describes a novel simulation approach that combines machine le...

Please sign up or login with your details

Forgot password? Click here to reset