Pagoda README

This README would normally document whatever steps are necessary to get your application up and running.

What is this repository for?

• Pagoda is one GPU runtime system which virtualizes GPU hardware resources to increase the utilization of GPU in narrow task workloads.
• 1.0

How do I get set up?

• Install CUDA library >= 7.5 and python >=2.6

• Configuration

  1. To change the value of SM_NUM in common/para.h based on your GPU.

  2. To change the path of CUDA and the value of sm (e.g. sm_35 or sm_52 ...) in common/make.config

• Dependencies

• Database configuration

• How to run tests

  1. To execute ppopp17 script after setting of the configuration

  2. The run script file in each benchmark folder

• Deployment instructions

Contribution guidelines

• Writing tests

  1. People can follow Pagoda APIs used in pagoda folders in each benchmark to work out your own pagoda programs.

• Code review

  1. baseline: this folder contains proglems composed by standard CUDA APIs

  2. pagoda : this folder includes programs that consist of Pagoda APIs

• Other guidelines

Who do I talk to?

• Contact: yeh14@purdue.edu