[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [GNUnet-developers] Faster (sequential ) inserting of records in MyS
Re: [GNUnet-developers] Faster (sequential ) inserting of records in MySQL.
Sun, 6 Apr 2003 00:11:03 +0300 (EEST)
On Sat, 5 Apr 2003, Christian Grothoff wrote:
> Personally, I think that while good performance is nice and all, users may be
> quite confused if the file that they put in does not instantly appear. Also,
> the implementation would get quite a bit more complicated, which is probably
> the best argument against it. After all, users are not going to insert
> content all the time; it's much more likely that they run gnunet-insert-multi
> once and are done with it (until we break the datastore format again... :-).
That batch-thing could be used to handle all arriving content, just
by appending them to the 'end of the file' that is regularly fed
to the db. In activemigrating system inserting is happening
all the time, its just not the local user that inserts.
But I'm not too keen on that kludge, but we might remember the
idea if db writes become a bottleneck, because basically Jan is
correct in the sense that it is reasonable to think it as a way
to make things more efficient on a mysql system. For other
dbmgrs, of course, its of less or no use.
ps. currently lookup of random content (for padding) seems to
cause mysql cpu peaks - I'm not exactly sure but I think I'm
seeing a correlation with randomlookup and occasional large
cpu usage. I'll look into it one of these days.