ASPP/2024-heraklion-parallel-python
8
0
Fork 21
Code Issues Pull requests 9 Actions Packages Projects Releases Wiki Activity

rewriting exercise

Browse source
This commit is contained in:
Jenni Rinker 2024-08-29 18:03:46 +03:00
parent 25dd96f746
commit 9bc9eb6292
1 changed files with 21 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

34
Exercise2a/README.md
View file

@ -1,44 +1,52 @@
# Exercise 2b
# Exercise 2a: multithreading with NumPy
Execute numpy code with multiple threads.
Objective: investigate speed-up of numpy code with multiple threads.
```NOTE``` Remember to use `htop` in your terminal to track what the CPUs are doing.
```HINT``` Use `htop` in your terminal to track what the CPUs are doing.
## First
Use the script `heavy_computation.py` with different numbers of threads.
The script `heavy_computation.py` performs some matrix calculations with numpy.
`OMP_NUM_THREADS` can be used to override the number of threads used:
You can change the number of threads that numpy uses for the calculation
using the `OMP_NUM_THREADS` environment variable like this:
```
OMP_NUM_THREADS=7 python heavy_computation.py
```
The script will save the timing results into a `timings/` folder as `.txt` files.
The script will also measure the time to run the calculation and will save
the timing results into the `timings/` folder as a `.txt` file.
**TASK**: Execute the script `heavy_computation.py`, varying the numbers of threads.
You will plot the resulting calculating times in the second part below.
**QUESTION**
> What happens if `OMP_NUM_THREADS` is not set? How many threads are there? Why?
## Second
Plot the timing results from the first part, we wrote the IO for you in `timing_plot.py`.
In `plot.py`, we have given code that will load all of the timing data in `timings/`.
1. Plot a graph of execution duration vs. the number of threads
2. Plot the execution speedup with respect to running a single-threaded process
**TASK**: Add code to plot of the execution duration vs. the number of threads
Open a PR with your plotting code and post your plots in the conversation, don't upload binaries to the Git remote!
**OPTIONAL TASK**: Add code to calculate and plot the speed-up time compared
to single-threaded execution. Include your code and plot in the PR.
**QUESTIONS**
> What does the result tell us about the optimum number of threads? Why?
> Does it take the same time as your colleagues to run? Why?
## Extra
## Optional tasks
Investigate the runtime variability. Systematically run multiple instances with the same number of threads by modifying `heavy_computation.py`.
### Extra extra
How is the runtime affected when the problem becomes bigger? Is the optimum number of threads always the same?
How is the runtime affected when the memory is almost full? You can fill it up by creating a separate (unused) large numpy array.
How about running on battery vs having your laptop plugged in?
How about running on battery vs. having your laptop plugged in?