Mat Object: 4 MPI processes
  type: mpisbaij
  rows=1000, cols=1000
  total: nonzeros=1999, allocated nonzeros=2000
  total number of mallocs used during MatSetValues calls=0
    [0] Local rows 250 nz 500 nz alloced 500 bs 1 mem 18900.
    [0] on-diagonal part: nz 250 
    [0] off-diagonal part: nz 250 
    [1] Local rows 250 nz 500 nz alloced 500 bs 1 mem 18900.
    [1] on-diagonal part: nz 250 
    [1] off-diagonal part: nz 250 
    [2] Local rows 250 nz 500 nz alloced 500 bs 1 mem 18900.
    [2] on-diagonal part: nz 250 
    [2] off-diagonal part: nz 250 
    [3] Local rows 250 nz 499 nz alloced 500 bs 1 mem 19398.
    [3] on-diagonal part: nz 499 
    [3] off-diagonal part: nz 0 
    Information on VecScatter used in matrix-vector product: 
    PetscSF Object: 4 MPI processes
      type: basic
        sort=rank-order
      [0] Number of roots=250, leaves=1, remote ranks=1
      [0] 0 <- (3,249)
      [1] Number of roots=250, leaves=1, remote ranks=1
      [1] 0 <- (3,249)
      [2] Number of roots=250, leaves=1, remote ranks=1
      [2] 0 <- (3,249)
      [3] Number of roots=250, leaves=0, remote ranks=0
      [0] Roots referenced by my leaves, by rank
      [0] 3: 1 edges
      [0]    0 <- 249
      [1] Roots referenced by my leaves, by rank
      [1] 3: 1 edges
      [1]    0 <- 249
      [2] Roots referenced by my leaves, by rank
      [2] 3: 1 edges
      [2]    0 <- 249
      [3] Roots referenced by my leaves, by rank