Optimal Hardware Parameters Prediction for Best Energy-to-Solution of Sparse Matrix Operations Using Machine Learning Techniques

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F18%3APU130709" target="_blank" >RIV/00216305:26230/18:PU130709 - isvavai.cz</a>
Result on the web
<a href="https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&ved=2ahUKEwj45dbYkMfdAhUilYsKHQe1DkwQFjADegQIBxAC&url=https%3A%2F%2Fwww.thinkmind.org%2Fdownload_full.php%3Finstance%3DINFOCOMP%2B2018&usg=AOvVaw0F5eFy3SoDGqt3wTWnO1GV" target="_blank" >https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=4&ved=2ahUKEwj45dbYkMfdAhUilYsKHQe1DkwQFjADegQIBxAC&url=https%3A%2F%2Fwww.thinkmind.org%2Fdownload_full.php%3Finstance%3DINFOCOMP%2B2018&usg=AOvVaw0F5eFy3SoDGqt3wTWnO1GV</a>
DOI - Digital Object Identifier
—

Result language
angličtina
Original language name
Optimal Hardware Parameters Prediction for Best Energy-to-Solution of Sparse Matrix Operations Using Machine Learning Techniques
Original language description
Combinations of 3 hardware parameters (number of threads, core and uncore frequency) were tested for each of the 4 sparse algorithms (matrix-matrix addition, matrix-matrix multiplication, matrix-vector multiplication in IJV and CSR format) on a set of several thousands matrices for the purpose of identifying the best energy-to-solution setting for each matrix and sparse operation. On this set of data, the possibility of optimal hardware setting prediction based on the properties of each matrix were analysed for each sparse algorithm. A calculation of Pearson correlation coefficient between the matrices' properties and optimal hardware parameters showed no direct correlation (highest 0.33 for x-y, lowest -0.25 for a-b). A neural network with back-propagation learning was used for deeper analysis to see if matrix properties correspond to hardware settings. The input neurons represented properties of given matrix, output neurons represented optimal hardware parameters. Network properties (hidden neurons per layer, hidden neuron layers, learning coefficient and learning strategy) impact on prediction accuracy were analysed and the results showed
Czech name
—
Czech description
—

Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)