jackrabbit-oak-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Ian Boston <...@tfd.co.uk>
Subject Re: Concurrency Question
Date Thu, 31 Jan 2013 09:15:29 GMT
On 31 January 2013 19:21, Thomas Mueller <mueller@adobe.com> wrote:
> Hi,
>>no, it's quite different. the equivalent to the journal in Jackrabbit 2.x
>>is rather the commit collection.
>>the sync collection contains only a single document, which points to
>>the head of the commit history. so, it is somewhat similar to the
>>InstanceRevision we have in Jackrabbit 2.x clustering.
> Sorry for the confusion, Marcel is right of course. The only way that the
> MongoMK "sync" collection is similar to the Jackrabbit 2.x journal is:
> write access to both is serialized. But that's it. (I hope I got that
> right now :-)

IIRC In JR2 each journal update was in a transaction that incremented
the revision number, so the sequence had no holes.

In MongoMK, am I right in thinking that since the sync collection only
points to the head revision it only needs to be locked for as long as
it takes to write the pointer to the head revision vastly reducing the
time which other writers are blocked.

Is that right or is the serialisation the same as in JR2 ?

Also wondering: Do you need a sync at all or would a fuzzy distributed
time signal (as used in Google Spanner) would give you enough padding
between operations to know their sequence ?
Please dont waste any time on this if its a simple "no".


> Regards,
> Thomas

View raw message