# KSPPIPECG
Pipelined conjugate gradient method. [](sec_pipelineksp) 


## Notes
This method has only a single non-blocking reduction per iteration, compared to 2 blocking for standard CG.  The
non-blocking reduction is overlapped by the matrix-vector product and preconditioner application.

See also `KSPPIPECR`, where the reduction is only overlapped with the matrix-vector product and `KSPGROPPCG`

MPI configuration may be necessary for reductions to make asynchronous progress, which is important for performance of pipelined methods.
See [](doc_faq_pipelined)


## Contributed by
Pieter Ghysels, Universiteit Antwerpen, Intel Exascience lab Flanders


## Reference
P. Ghysels and W. Vanroose, "Hiding global synchronization latency in the preconditioned Conjugate Gradient algorithm",
Submitted to Parallel Computing, 2012.


## See Also
 [](chapter_ksp), [](doc_faq_pipelined), [](sec_pipelineksp), `KSPCreate()`, `KSPSetType()`, `KSPPIPECG2`, `KSPPIPECR`, `KSPGROPPCG`, `KSPPGMRES`, `KSPCG`, `KSPCGUseSingleReduction()`

## Level
intermediate

## Location
<A HREF="PETSC_DOC_OUT_ROOT_PLACEHOLDER/src/ksp/ksp/impls/cg/pipecg/pipecg.c.html#KSPPIPECG">src/ksp/ksp/impls/cg/pipecg/pipecg.c</A>


---
[Edit on GitLab](https://gitlab.com/petsc/petsc/-/edit/release/src/ksp/ksp/impls/cg/pipecg/pipecg.c)


[Index of all KSP routines](index.md)  
[Table of Contents for all manual pages](/docs/manualpages/index.md)  
[Index of all manual pages](/docs/manualpages/singleindex.md)  
