search

GraphBLAS / LAGraph

This is a library plus a test harness for collecting algorithms that use the GraphBLAS. For test coverage reports, see https://graphblas.org/LAGraph/ . Documentation: https://lagraph.readthedocs.org
Other
232 stars 61 forks source link

Recent algorithms for triangle counting #29

Open szarnyasg opened 5 years ago

szarnyasg commented 5 years ago

I've recently read the HPEC 2017 paper on triangle counting without matrix multiplication co-authored by @mcmillan03 and skimmed through the followup Correctness 2017 paper. I've found the GBTL implementation of the algorithms presented in the latter: https://github.com/cmu-sei/gbtl/blob/master/src/algorithms/triangle_count.hpp

It seems to me that these could be ported to SuiteSparse:GraphBLAS. Would it be an interesting line of work to port these algorithms? Can we make any predictions regarding the performance of these algorithms in SuiteSparse?

mcmillan03 commented 5 years ago

I have ported algorithms from GBTL to C API an number of times and it is relatively straightforward. I have a task on my backlog to do this for LAGraph. Just need to find time.

On Thu, Oct 17, 2019 at 11:36 PM Gabor Szarnyas notifications@github.com wrote:

I've recently read the HPEC 2017 paper on triangle counting without matrix multiplication https://users.ece.cmu.edu/~franzf/papers/hpec_2017_low.pdf co-authored by @mcmillan03 https://github.com/mcmillan03 and skimmed through the followup Correctness 2017 paper https://dl.acm.org/citation.cfm?id=3145484. I've found the GBTL implementation of the algorithms presented in the latter: https://github.com/cmu-sei/gbtl/blob/master/src/algorithms/triangle_count.hpp

It seems to me that these could be ported to SuiteSparse:GraphBLAS. Would it be an interesting line of work to port these algorithms? Can we make any predictions regarding the performance of these algorithms in SuiteSparse?

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/GraphBLAS/LAGraph/issues/29?email_source=notifications&email_token=AANXEP7UWVHPFAA6UXH6YJ3QPFKOBA5CNFSM4JCDFNG2YY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4HSU6J4A, or unsubscribe https://github.com/notifications/unsubscribe-auth/AANXEP4M3RYISSOO3MDPGMTQPFKOBANCNFSM4JCDFNGQ .