167
edits
Changes
→Resources
This is our main idea to parallel the grep function. Yes, it still have some loops. But those loops are not as large as that in serial grep function. So we start to design the code foe it.<br>
''' Parallel grep code '''
Here are part of parallel grep code:<br>
Main grep parallel:<br>
[[File:ParaGrep.JPG|650px]]
Initial integer arrar kernel:<br>
[[File:InitialBandFunc.JPG|650px]]
Grep character kernel:<br>
[[File:GrepKernel.JPG|650px]]
Two grep:<br>
[[File:GrepCompare.JPG|650px]]
Run and result:<br>
[[File:Result22.JPG|650px]]
Yes. The grep of parallel's result is the same as serial's result! It works!<br>
''' Profile '''
Sample of profile:<br>
[[File:ProfilePara.JPG|650px]]
Run and record time data for serial program and parallel program. We got the time for each:<br>
[[File:TimeCompareTwoPart.JPG|650px]]
Generate the graphics:<br>
Total time compare:<br>
[[File:TotalCompare.JPG|650px]]
Time compare for function grep() and grepPara():<br>
[[File:TimecompareFunction.JPG|650px]]
We got it!<br>
=== Assignment 3 ===
Next, we optimize the code so that it runs faster.<br>
''' Use shared memory '''
We change global memory to shared memory in kernel grepx(). The code is:<br>
[[File:SharedMem.JPG|650px]]
''' Profile '''
We get the data like this:<br>
[[File:SharedData.JPG|650px]]
''' ntpb ''
We also try different ntpb and see if it effect runtime.<br>
[[File:NtpbDiff.JPG|650px]]
According to the data, the number of ntpb is not an important factor affecting time.
''' End '''
That's all for our assignment.
== Resources ==
You can see my code here: https://gist.github.com/KignorChan/764c71d9d96561e4101d9830e0bc5e69
Idea from LZW compress program: https://codereview.stackexchange.com/questions/86543/simple-lzw-compression-algorithm
