﻿<?xml version="1.0" encoding="utf-8"?><rss version="2.0"><channel><title>Ayende @ Rahien</title><link>http://ayende.com</link><description>Ayende @ Rahien</description><copyright>Copyright (C) Ayende Rahien  2004 - 2021 (c) 2026</copyright><ttl>60</ttl><item><title>Ayende Rahien commented on RavenDB: Self optimizing Ids</title><description>Jesse,
We have this option, sure. See here:
http://ravendb.net/docs/theory/document-key-generation</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment15</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment15</guid><pubDate>Tue, 24 Apr 2012 16:42:45 GMT</pubDate></item><item><title>Jesse commented on RavenDB: Self optimizing Ids</title><description>Thank you for the fast reply! I understand that this could be handy for bulk inserts, but wouldn't the majority of inserts be 1 at a time? It's pretty annoying to have my Id's be 1, 32, 64, etc. And maybe I just need to look into this more before getting too frustrated, and maybe there's a simply way to do something like "id = getmaxid +1"</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment14</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment14</guid><pubDate>Tue, 24 Apr 2012 16:19:02 GMT</pubDate></item><item><title>Ayende Rahien commented on RavenDB: Self optimizing Ids</title><description>Jesse,
Huh? How did you get to that?
What we are actually doing is optimizing ourselves on demand, so if you are inserting bulk data, we would reduce the number of requests to the server</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment13</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment13</guid><pubDate>Tue, 24 Apr 2012 15:51:28 GMT</pubDate></item><item><title>Jesse commented on RavenDB: Self optimizing Ids</title><description>So basically, you have to insert 32 documents at a time, and that is the only option other than looking at the last Id, and manually updating the Id integer for every document? Or you can query the whole collection for the last Id.. that's gotta be a better way.</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment12</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment12</guid><pubDate>Tue, 24 Apr 2012 15:34:03 GMT</pubDate></item><item><title>Judah Gabriel Himango commented on RavenDB: Self optimizing Ids</title><description>Wow, Raven learns from recent usage patterns. Very cool.</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment11</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment11</guid><pubDate>Tue, 27 Mar 2012 20:56:53 GMT</pubDate></item><item><title>Edward Spelt commented on RavenDB: Self optimizing Ids</title><description>Mauro, I do not (yet) have replication setups with RavenDB.

Can a slave in RavenDB also do CUD?

At this moment we have a few MSSQL DB's which have merge replication.   These are master/master../master types which can also work in a disconnected enviroment (nice with crappy connections) and thus on different physical locations.
Althought it works nice it is far from zero config and sometimes high maintenance. 

</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment10</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment10</guid><pubDate>Tue, 27 Mar 2012 13:58:38 GMT</pubDate></item><item><title>Mauro commented on RavenDB: Self optimizing Ids</title><description>Edward, I think it's a bad idea to have a multi-master db (where multi &gt;2), I'm trying a master-multislave configuration with failover instead</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment9</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment9</guid><pubDate>Tue, 27 Mar 2012 12:31:44 GMT</pubDate></item><item><title>Ayende Rahien commented on RavenDB: Self optimizing Ids</title><description>Edward,
Yes, that is the idea. Or you use a guid, but that isn't as readable.</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment8</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment8</guid><pubDate>Tue, 27 Mar 2012 11:13:51 GMT</pubDate></item><item><title>Edward commented on RavenDB: Self optimizing Ids</title><description>Its a master/master, could also be a master/master/master.

You mean users/serverId/1234  ?</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment7</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment7</guid><pubDate>Tue, 27 Mar 2012 10:15:39 GMT</pubDate></item><item><title>Ayende Rahien commented on RavenDB: Self optimizing Ids</title><description>Edward,
Is it a master / master? Master / slave?
If it is master slave, it is easy.

If it is master / master, you probably was to add the write server id to the id, to ensure uniqueness, something like users/1/1234</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment6</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment6</guid><pubDate>Tue, 27 Mar 2012 10:08:56 GMT</pubDate></item><item><title>Edward commented on RavenDB: Self optimizing Ids</title><description>What would be the recommended way for generating human readable id's for a replicated enviroment?</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment5</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment5</guid><pubDate>Tue, 27 Mar 2012 09:32:58 GMT</pubDate></item><item><title>Ayende Rahien commented on RavenDB: Self optimizing Ids</title><description>Edward,
You generally do not generate ids in this manner in a replicated env.</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment4</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment4</guid><pubDate>Tue, 27 Mar 2012 08:56:17 GMT</pubDate></item><item><title>Edward commented on RavenDB: Self optimizing Ids</title><description>Does this generation of id's not conflict in a replicated enviroment?</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment3</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment3</guid><pubDate>Mon, 26 Mar 2012 17:43:58 GMT</pubDate></item><item><title>Ayende Rahien commented on RavenDB: Self optimizing Ids</title><description>Shane,
Guid aren't human readable.</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment2</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment2</guid><pubDate>Mon, 26 Mar 2012 14:56:05 GMT</pubDate></item><item><title>Shane Courtrille commented on RavenDB: Self optimizing Ids</title><description>What was the reason you didn't go with Comb Guid which wouldn't have the same problem?  Performance of some kind or just personal preference?</description><link>http://ayende.com/154337/ravendb-self-optimizing-ids#comment1</link><guid>http://ayende.com/154337/ravendb-self-optimizing-ids#comment1</guid><pubDate>Mon, 26 Mar 2012 14:52:07 GMT</pubDate></item></channel></rss>