On 08/03/11 10:37, [email protected] wrote:
Hi,
From: [email protected] [[email protected]] on
behalf of ext Clark, Joel [[email protected]] Sent: 08 March 2011
12:29
Why would you think MeeGo only runs on a phone?
And even if it was a phone, considering that nowadays there are tens
or hundreds of GB of storage, as business user I'd rather play it
safe and have the whole directory stored locally.
Typically one ends up in a foreign country with roaming problems,
barely compatible/reliable phone connection, no WiFi and need to
call/SMS someone.
All of this without taking into account roaming costs that might
actually be prohibitive for some.
Right, but disk space is not the issue here. As said before, querying is
what we optimise for primarily, if you want to insert or merge 120k
contacts, that's ok, but don't expect it to be super fast.
Before we had SPARQL, we had equivalent speed for updating and querying
and the speed for querying was not good enough. So we focused on
querying because that's the primary use case for every target that
Tracker is used on. The frequency of syncing or updating data in general
is much less common than querying it.
To make querying faster, it is a careful balance between data redundancy
in tables (to avoid joining with other tables) and indexes on columns
(to make lookups faster and avoiding full table scans). Adding this data
and these indexes increases disk space and indexing times but increases
query times as a direct result.
As you can imagine, if you want to be able to query, insert and merge
120k of contracts, you really can't have your cake and eat it so to
speak without some sacrifices one way or another.
As Philip says, there are still some areas to improve on and we're
working on that.
--
Regards,
Martyn
_______________________________________________
MeeGo-dev mailing list
[email protected]
http://lists.meego.com/listinfo/meego-dev
http://wiki.meego.com/Mailing_list_guidelines
