CS 378: Programming for Performance

Assignment 1: Performance counters

Due date: February 7

Late submission policy: Submission can be at the most 2 days late. There will be a 10% penalty for each day after the due date (cumulative).

Description

Write C code for the 6 variants of matrix-matrix multiply (matrixC = matrixA * matrixB) you can generate by permuting loops. The data type in the matrix should be doubles.

Instrument your implementations using PAPI to measure:
- Total Load Store Instructions
- Total Floating Point Instructions
- Data L1 miss
- Total L2 miss
Compile your code in ICC with flags '-O3 -fp-model precise'.
Using stampede at TACC, collect these measurements for two matrix sizes: 200 and 2000. Since the values you obtain will depend a lot on the machine you use, you must use stampede for the numbers you report. To ensure no interference with other processes, submit your runs to the job scheduler - use the 'serial' queue.
For each of the matrix sizes, plot the L1 and L2 miss rates as well as the total load and store instructions and floating point instructions of each loop-order variant.
For the smaller matrix size, does the L1 and L2 miss rates vary for the different loop-order variants? Does it vary for the larger matrix size? Is there any difference in behavior between the different problem sizes? Can you explain the reason for all of these answers?
Hint: To check cache sizes on the machine, run:
```
lscpu
```

Deliverables

Submit (in canvas) your code and a PDF containing the plots and explanation.

Grading

Code: 40 points

Measurements (plots): 40 points

Explanation: 20 points

Stampede on TACC:

Use the login node only for development - do not run or debug any executable on it. Run and debug your applications using the job scheduler.

Read http://www.tacc.utexas.edu/user-services/user-guides/stampede-user-guide#running to learn how to submit jobs.

PAPI:

To program with PAPI on stampede, run:

module load papi

For help on using the module, run:

module help papi

For more information on using modules, check https://portal.tacc.utexas.edu/user-guides/stampede#compenv-modules

To see which papi counters are available on a host, run:

papi_avail

Read the PAPI manual http://icl.cs.utk.edu/projects/papi/wiki/PAPIC:EventSets for more information, including example code.

"Warning! num_cntrs is more than num_mpx_cntrs" can be ignored.