Scalable iterative solution of sparse linear systems
Parallel Computing
ILUM: a multi-elimination ILU preconditioner for general sparse matrices
SIAM Journal on Scientific Computing
Graph mining: Laws, generators, and algorithms
ACM Computing Surveys (CSUR)
A framework for scalable greedy coloring on distributed-memory parallel computers
Journal of Parallel and Distributed Computing
SIAM Journal on Scientific Computing
A new scalable parallel DBSCAN algorithm using the disjoint-set data structure
SC '12 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
ColPack: Software for graph coloring and related problems in scientific computing
ACM Transactions on Mathematical Software (TOMS)
| Hi-index | 0.00 |
We present new multithreaded vertex ordering and distance-k graph coloring algorithms that are well-suited for multicore platforms. The vertex ordering techniques rely on various notions of "degree", are known to be effective in reducing the number of colors used by a greedy coloring algorithm, and are generic enough to be applicable to contexts other than coloring. We employ approximate degree computation in the ordering algorithms and speculation and iteration in the coloring algorithms as our primary tools for breaking sequentiality and achieving effective parallelization. The algorithms have been implemented using OpenMP, and experiments conducted on Intel Nehalem and other multicore machines using various types of graphs attest that the algorithms provide scalable runtime performance. The number of colors the algorithms use is often close to optimal. The techniques used for computing the ordering and coloring in parallel are applicable to other problems where there is an inherent ordering to the computations that needs to be relaxed for increasing concurrency.