First day of the fifth week I found out that I could play around by changing the number of blocks and threads that I was using to see if the speed of the program would change.
Some times using a higher number of blocks and a lower number of threads would make the program work slower, having a low number of blocks and high number of threads would at leas make the program work faster than having more blocks with less threads.
I was assigned my second program called Barnes Hut.
Understanding what the Barnes Hut program is about.
The program was not compiling before the parallelization this kept me being able to parallelize the program.
I was told to find a working version of this program on the internet this was a major problem I used many different versions of this program but I was not successful.
Due to this problem I was not really productive this week, instead I looked to the next assigned program. This contained some issues too which I was not able to identify.
By the end of the week I read about more OpenCV syntax. This was useful to start the parallelization of the program that will need to be presented at the XSEDE conference.
Some times using a higher number of blocks and a lower number of threads would make the program work slower, having a low number of blocks and high number of threads would at leas make the program work faster than having more blocks with less threads.
I was assigned my second program called Barnes Hut.
Understanding what the Barnes Hut program is about.
- Is an algorithm that divides a set of bodied into groups
- It uses a quad-tree
- The nodes represent a region of the two dimensional spac
- The nodes are divided into two types internal and external.
External Nodes:- has no children
- Empty or is a single body
- Represents the group of bodies under it.
- stores center of mass of its children
- total mass of its children
- It calculates the forces acting on a body
- Calculates Net force
- simulates gravitational force between two bodies
- has no children
The program was not compiling before the parallelization this kept me being able to parallelize the program.
I was told to find a working version of this program on the internet this was a major problem I used many different versions of this program but I was not successful.
Due to this problem I was not really productive this week, instead I looked to the next assigned program. This contained some issues too which I was not able to identify.
By the end of the week I read about more OpenCV syntax. This was useful to start the parallelization of the program that will need to be presented at the XSEDE conference.