Neil Carlson wrote:
>compute something like the md5 checksum of the data (these would be
>Fortran arrays, not a disk file)
>
sounds like a fine plan to me. You can probably use any of a wide
variety of hashing/signature functions. Depending on how often you need
to recompute the signature might motivate a search for a simpler/faster
algorithm with acceptable accuracy for your task (I presume that a false
positive (trigger a recompute when one wasn't necessary) is much less of
a problem than a false negative (use the cached result when you ought
not ;>). Also, if the values are floating point, if you require strict
bit for bit equivalence vs. some sort of approximate equivalence (on
some processors, notably x86 ... the magic of 80-bit intermediates might
make this a bit tricker question than it seems at first blush).
>and include it with the cached
>results. Then all I need to do the compare md5 checksums. Is this
>a reasonable plan? Is there something better? What about a free
>fortran implementation of md5 checksums?
>
>
>
I don't think I've ever seen md5 implemented in Fortran.
--
Keith H. Bierman [log in to unmask]
Sun Microsystems PAE | [log in to unmask]
15 Network Circle UMPK 12-325 | 650-352-4432 voice+fax
Menlo Park, California 94025 | sun internal 68207
<speaking for myself, not Sun*> Copyright 2004
------------------------------------------------------------
NOTICE: This email message is for the sole use of the
intended recipient(s) and may contain confidential and
privileged information. Any unauthorized review, use,
disclosure or distribution is prohibited.
If you are not the intended recipient, please contact
the sender by reply email and destroy all copies of the
original message.
|