Interactive visualization & performance benchmarks
LU decomposition factors a matrix A into the product of a lower triangular matrix (L) and an upper triangular matrix (U). This is useful for solving systems of linear equations, computing determinants, and matrix inversion.
Real benchmark results comparing Sequential, OpenMP, and Pthreads implementations on various matrix sizes and thread counts. Tests performed on actual hardware to measure parallelization efficiency.
| Matrix Size | Sequential | 1 Thread | 2 Threads | 4 Threads | 8 Threads | 16 Threads |
|---|---|---|---|---|---|---|
| 500×500 | 0.0255 | 0.0141 | 0.0082 | 0.0061 | 0.0066 | 0.0307 |
| 1000×1000 | 0.2021 | 0.1065 | 0.0558 | 0.0471 | 0.0327 | 0.0612 |
| 1500×1500 | 1.1206 | 0.7358 | 0.3470 | 0.2036 | 0.2868 | 0.3013 |
| 2000×2000 | 2.9234 | 2.7630 | 2.0936 | 1.8705 | 1.6748 | 2.5432 |
| Matrix Size | Sequential | 1 Thread | 2 Threads | 4 Threads | 8 Threads | 16 Threads |
|---|---|---|---|---|---|---|
| 500×500 | 0.0255 | 0.0118 | 0.0278 | 0.0496 | 0.0875 | 0.1669 |
| 1000×1000 | 0.2021 | 0.0990 | 0.0920 | 0.1188 | 0.2212 | 0.3476 |
| 1500×1500 | 1.1206 | 0.7468 | 0.4443 | 0.3638 | 0.5221 | 0.7597 |
| 2000×2000 | 2.9234 | 2.5724 | 1.4303 | 1.4503 | 1.6226 | 1.9806 |