Skip to content

Project done as a course-project for CS422 (Computer Architecture) at IIT Kanpur, in April-March 2015, under guidance of prof. Mainak Chaudhuri. Idea is to implement the popular perceptron based branch predictors for different memory limits and evaluate the performance on CBP-2014 kit. Also try to explain the performance result.

Notifications You must be signed in to change notification settings

vickianand/perceptron-based-branch-predictors

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Setting up the CBP-4 infrastructure and simulating the given predictors

  • The download file and setting-up instruction for CBP-4 Kit is available at site : http://www.jilp.org/cbp2014/framework.html
  • There are predictor.cc and predictor.h files in all three directories - gshare, perceptron, and piecewise_linear.
  • For simulating the given predictors, you'll need to replace the predictor.cc and predictor.h file in cbp2014/sim/ directory by those from the directory of the required predictor.
  • In cbp2014/sim/ direcory execute : rm *.o and rm predictor followed by make. This will create the executable of the required predictor.
  • After this, for running the predictors and generating results, the instructions given on the CBP-4 site should be followed.

Changing the budgets for the predictors

  • The storage budgets for the both the predictors can be changed in respective .h files by changing the #define Budget 4 line.
  • The budgets for the perceptron predictor can be set to 4 or 8 or 32.
  • The budgets for the piece-wise linear predictor can be set to 4 or 8 or 32 or 256.
  • All the above budgets represent the budget in KB.
  • The required calculations as per the budget for different parameters have been shown in comments alongside the initializations of the parameters.
  • The included gshare directory is the one provided by default with the cbp2014 kit.

About

Project done as a course-project for CS422 (Computer Architecture) at IIT Kanpur, in April-March 2015, under guidance of prof. Mainak Chaudhuri. Idea is to implement the popular perceptron based branch predictors for different memory limits and evaluate the performance on CBP-2014 kit. Also try to explain the performance result.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages