I have an unformatted file, and I open it for reading in big
endian mode. The file has a repetitive structure (is a time step
report for a simulation): at the beginning of every time step, there
is this pseudo-declaration:

'PARAMS'
keyword number kind

where "keyword" is an 8 character string, "number" is an integer which
tells me how much data this time step contains, and "kind" is a 4
character string which tells me the data type (i.e., 'CHAR', 'REAL'
and so on). After this declaration, the data for the current time step
follows. As an example:

'PARAMS  '  4200 'REAL'
0.5372 3.7393 0.3273 .... (4200 real numbers)

With the file I am working on, this 4200 stays constant for every time
step in the simulation.
The numerical arrays are divided into blocks of up to 1000 items each:
so, in my previous example, I suppose I can not read the whole 4200
real numbers in one shot, I have to loop 4 times and read 1000 items
every time, plus read the remaining 200 as last step.

I have written the following routine to read and store the data from
this file. Is there anything I could do to speed up the reading/
storing of the data?

Thank you very much for your suggestions and comments. I apologize for
the long post.

Andrea.

! Begin Code

subroutine ReadSMSPEC(fileName, miniSteps, dimens, matrix)

        ! PARAMETERS:
        ! fileName: the file containing the data
        ! miniSteps: the number of time steps in the simulation
        ! dimens: the number of data in each time step
        ! matrix: a dimens x miniStep 2D array of real data

        character*1000, intent(in) :: fileName
        integer, intent(in) :: miniSteps, dimens
        real(4), intent(out) :: matrix(dimens, miniSteps)

        integer keywordNumber, internalCount, readBuffer, matrixIndex

        character*8 keywordName
        character*4 keyworType

        logical feof

        ! Open the file
        open(unit=1, file=fileName, form='UNFORMATTED', convert='BIG_ENDIAN')

        feof = .false.

        ! The numerical arrays are divided into blocks of up to 1000 items
each
        readBuffer = int(dimens/1000)

        do while (.not.feof)

                ! Read the keyword, number and kind
            read(1, end=40) keywordName, keywordNumber, keywordType

                ! I look for the 'PARAMS' keyword
            if (keywordName == 'PARAMS') then

                        ! A new time step has been found
                        ! Increase the index in the matrix
                        matrixIndex = matrixIndex + 1

                    internalCount = 0

                        ! Loop by reading a block of 1000 items
                        ! at a time

                        do while (internalCount <= readBuffer)

                                ! Read a block of 1000 items until
                                ! you reach the end of this time step

                                if (internalCount < readBuffer) then

                                        ! We are still far from the last block,
                                        ! Read 1000 items of data

                            read(1, end=40) matrix(1000*internalCount
+1:1000*(internalCount+1), matrixIndex)

                        else

                                        ! We reached the last block of data:
                                        ! Read the remaining data, which can be
                                        ! less than 1000 items.

                                        read(1, end=40) matrix(1000*internalCount+1:dimens, matrixIndex)

                                endif

                                internalCount = internalCount + 1

                        enddo

                endif

        enddo

40      continue

        close(1)
        return

end subroutine readSMSPEC