It probably won't meet their needs either, but the latest version of DirectoryStorage is supposed to be compatible with Zope3.<br>It might mitigate the startup time and index memory consumption issues somewhat, especially if they can use reiserfs.
<br><br><a href="http://dirstorage.sourceforge.net/">http://dirstorage.sourceforge.net/</a><br><br>Matt<br><br><div><span class="gmail_quote">On 1/22/06, <b class="gmail_sendername">Jeff Rush</b> <<a href="mailto:jeff@taupro.com">
jeff@taupro.com</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">David Johnson wrote:<br>><br>> In general we're looking at using an RDBMS, but we're trying to
<br> > get a better feeling of when we can use the ZODB and when we<br> > should stick to the RDBMS.<br>><br>> We have various datasets:<br>><br>> 1. Transactions: 10-100 billion objects<br>> 2. Users: 10-100 million objects
<br>> 3. Contacts: 100,000 objects<br>> 4. Customers: 100-1000 objects<br><br>I see other good replies but one factor you must keep in mind for ZODB is<br>that it keeps the index of all objects in memory. With 100 billion
<br>transaction objects, you're going to need a LOT of RAM and have a large<br>startup-time.<br><br>I've always wondered why ZODB didn't offer an option of storing the object<br>index in a BTREE, but I guess its the performance issue of potential
<br>additional disk accesses.<br><br>-Jeff<br><br>_______________________________________________<br>Zope3-users mailing list<br><a href="mailto:Zope3-users@zope.org">Zope3-users@zope.org</a><br><a href="http://mail.zope.org/mailman/listinfo/zope3-users">
http://mail.zope.org/mailman/listinfo/zope3-users</a><br></blockquote></div><br>