Sparse Data Patterns

Coordinate Row Sum

Grouping Sparse Entries

Coordinate sparse storage keeps row, column, and value arrays side by side.

Program

Play the program to sum entries from a different sparse row.

coordinate_row_sum.f90 
program coordinate_row_sum_demo
    implicit none
    integer :: rows(4)
    integer :: cols(4)
    integer :: values(4)
    integer :: target_row
    integer :: i
    integer :: count
    integer :: total

    rows = [1, 2, 2, 3]
    cols = [1, 1, 3, 2]
    values = [5, 7, 4, 6]
    target_row = 
    count = 0
    total = 0
    do i = 1, 4
        if (rows(i) == target_row) then
            count = count + 1
            total = total + values(i)
        end if
    end do
    print '(I0, 1X, I0, 1X, I0)', target_row, count, total
end program coordinate_row_sum_demo

program coordinate_row_sum_demo
    implicit none
    integer :: rows(4)
    integer :: cols(4)
    integer :: values(4)
    integer :: target_row
    integer :: i
    integer :: count
    integer :: total

    rows = [1, 2, 2, 3]
    cols = [1, 1, 3, 2]
    values = [5, 7, 4, 6]
    target_row = 
    count = 0
    total = 0
    do i = 1, 4
        if (rows(i) == target_row) then
            count = count + 1
            total = total + values(i)
        end if
    end do
    print '(I0, 1X, I0, 1X, I0)', target_row, count, total
end program coordinate_row_sum_demo

program coordinate_row_sum_demo
    implicit none
    integer :: rows(4)
    integer :: cols(4)
    integer :: values(4)
    integer :: target_row
    integer :: i
    integer :: count
    integer :: total

    rows = [1, 2, 2, 3]
    cols = [1, 1, 3, 2]
    values = [5, 7, 4, 6]
    target_row = 
    count = 0
    total = 0
    do i = 1, 4
        if (rows(i) == target_row) then
            count = count + 1
            total = total + values(i)
        end if
    end do
    print '(I0, 1X, I0, 1X, I0)', target_row, count, total
end program coordinate_row_sum_demo
coordinate arrays `rows`, `cols`, and `values` describe each stored entry.
row filter Only entries whose row matches `target_row` contribute to the summary.
sparse row total `count` and `total` describe the selected sparse row.