thanks for your feedback. In
you find Rule 7:
If a tool other than the included pfind is used for the find phase,
then it must follow the same input and output behavior as the included pfind and the
source code must be included in the submission.
based on your comment, the rule
might need some clarification.
IMHO the intention is that everyone will be able to use the tool to
reproduce your results (and not having some "magic" result).
A proprietary tool that is nowhere distributed (e.g., written by you
or a company) would be problematic.
I think a vendor ensures that their great tool is somehow available to
their customers at least.
A logical approach to the rule could be:
if the tool is closed source (but typically available on the storage
system you use), you should include a wrapper file in the submission
like the sfind wrapper:
If you make a modification to an existing open-source find, then you
must include the whole source code (such as a copy of the modified
Maybe the committee will discuss this and clarify the rules.
P.S. These statements merely reflect my own personal view; the only
mechanism for announcing official IO500 policies and decisions is the
committee(a)io500.org email address.
On Mon, Jun 1, 2020 at 7:17 PM Carlile, Ken via IO-500 <io-500(a)vi4io.org> wrote:
Hi all (and Committee in particular),
Is it permissible to use an external piece of software for the find portion wrapped by
some code to make it accept the find-style parameters (-name, -size, -newer), and if so,
what code would need to be submitted?
I am looking at using Starfish to do the find via doing a scan and then a query from
within a wrapper script that the io500 script would call. I believe at least some of the
Starfish software is closed source, so I obviously wouldn't be able to submit their
Sr. Storage Engineer
HHMI/Janelia Research Campus
IO-500 mailing list
Dr. Julian Kunkel
Lecturer, Department of Computer Science
+44 (0) 118 378 8218
PGP Fingerprint: 1468 1A86 A908 D77E B40F 45D6 2B15 73A5 9D39 A28E