# Solving independent problems in parallel **URL:** \ **Category:** General\ **Created:** [January 12, 2024, 7:53am UTC](https://fenicsproject.discourse.group/t/solving-independent-problems-in-parallel/13298 "2024-01-12T07:53:47Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![federicobetti](https://avatars.discourse-cdn.com/v4/letter/f/50afbb/32.png) [@federicobetti](https://fenicsproject.discourse.group/u/federicobetti)\ **Post date:** [January 12, 2024, 7:53am UTC](https://fenicsproject.discourse.group/t/solving-independent-problems-in-parallel/13298/1 "2024-01-12T07:53:47Z")
Hello, I am trying to solve independent problems in parallel. Up to 16 procs the speedup is good with respect to solving only 1 problem in serial, but then suddenly this drops dramatically. Inspecting the times, this slowdown is due to a slower assemble of the stiffness matrix. The solve time is constant on the other hand. Is there any reason/fix for this behaviour? Thanks, Federico ---
**Author:** ![federicobetti](https://avatars.discourse-cdn.com/v4/letter/f/50afbb/32.png) [@federicobetti](https://fenicsproject.discourse.group/u/federicobetti)\ **Post date:** [January 12, 2024, 7:59am UTC](https://fenicsproject.discourse.group/t/solving-independent-problems-in-parallel/13298/2 "2024-01-12T07:59:02Z")
Let me maybe ask a more specific question: the communicator of a matrix assembled with `dolfinx.fem.petsc.assemble_matrix(form, bcs)` is derived from the mesh one (which is MPI.COMM\_SELF) or not? If not, how can one achieve this? ---
**Author:** ![dokken](https://yyz2.discourse-cdn.com/free1/user_avatar/fenicsproject.discourse.group/dokken/32/1560_2.png) [@dokken](https://fenicsproject.discourse.group/u/dokken)\ **Post date:** [January 12, 2024, 9:35am UTC](https://fenicsproject.discourse.group/t/solving-independent-problems-in-parallel/13298/3 "2024-01-12T09:35:17Z")
The matrix assembler is derived from the one in the mesh. Without an example at hand, it is hard to pinpoint why your problem doesn’t scale after 16 processors.