Running TAP/async jobs immediately: Proposal Text
Patrick Dowler
pdowler.cadc at gmail.com
Fri Sep 30 21:45:50 CEST 2016
If people are doing quick queries that respond immediately and they
want to minimise connection overhead... TAP-sync? That's what it is
for. With TAP-async you have to stage the results someplace, which
adds way more latency than an extra connection... so, use TAP-sync!
While TAP-async could mandate support for this, we cannot remove
support for the multi-request UWS way (so you need that too) and thus
we can't mandate clients use this... which means Brian is right: you
can do this and your own client/portal can take advantage of it with
no spec change.
Pat
On 29 September 2016 at 10:33, Graham, Matthew J. <mjg at caltech.edu> wrote:
> Hi Walter,
>
> The IVOA documents page indicates that UWS 1.1 is under RFC and if you follow the link through to the actual RFC page, you’ll see that this ended last year. As mentioned in Stellenbosch, this version has been approved by all WG chair and is about to go to Exec for approval. Any further changes need to go into v1.2.
>
> Cheers,
>
> Matthew
>
>
>> On Sep 29, 2016, at 9:34 AM, Walter Landry <wlandry at caltech.edu> wrote:
>>
>> Mark Taylor <M.B.Taylor at bristol.ac.uk> wrote:
>>> Adjusting UWS to get off the fence about this would also be
>>> a possibility (though given that UWS is not up for change right
>>> now, not such a grat idea).
>>
>> I am confused. At
>>
>> http://www.ivoa.net/documents/UWS/
>>
>> the document is marked as a "Proposed Recommendation made available
>> for public review". The standards process is still not entirely clear
>> to me, but I thought that meant it is up for change right now.
>>
>>> But, it's a change to the current standard, which is always slightly
>>> painful. And to argue for it on the grounds of N vs N+1 (N>=3)
>>> connections seems a bit disingenuous, since if you want a quick in,
>>> quick out job you will probably be using the sync endpoint.
>>
>> I was thinking of jobs that will usually be fast, but might take
>> arbitrarily long. For example, interfaces that allow users to enter
>> arbitrary SQL. For what it is worth, looking around at existing user
>> friendly interfaces for async jobs (Gaia, TAPHandle, TOPCAT), it seems
>> that submitting a jobs always runs it immediately.
>>
>> Cheers,
>> Walter Landry
>
--
Patrick Dowler
Canadian Astronomy Data Centre
Victoria, BC, Canada
More information about the grid
mailing list