http://thirdrail.disqus.com/enMon, 06 Apr 2009 10:01:40 -0000http://3.rdrail.net/blog/announcing-thrudb-document-oriented-database-services/#comment-7902690<p>Thanks for bringing this up, just exactly what I needed, hope I can give some feedbacks after I checked it out.</p>electronic discoveryMon, 06 Apr 2009 10:01:40 -0000http://3.rdrail.net/blog/announcing-thrudb-document-oriented-database-services/#comment-6858765<p>Cool stuff. Haven't had a chance to dive into the code yet but I'd say you are onto something good.</p><p>The background you provide about databases, Google's infrastructure, etc., are spot on.</p><p>Here is a panel I participated in on this particular topic: <a href="http://www.geeksessions.com/2007/10/08/geeksessions-12-archived-episdes/" rel="nofollow noopener" target="_blank" title="http://www.geeksessions.com/2007/10/08/geeksessions-12-archived-episdes/">http://www.geeksessions.com...</a></p><p>Not sure if you are aware of Hadoop (<a href="http://lucene.apache.org/hadoop)" rel="nofollow noopener" target="_blank" title="http://lucene.apache.org/hadoop)">http://lucene.apache.org/ha...</a>, which is an open-source implementation of a large-scale distributed file system and map-reduce system. It is led by Doug Cutting, creator of Lucene, and Yahoo! has a pretty large group of people contributing to it. There is a sub-project of Hadoop called Hbase (<a href="http://wiki.apache.org/lucene-hadoop/Hbase)" rel="nofollow noopener" target="_blank" title="http://wiki.apache.org/lucene-hadoop/Hbase)">http://wiki.apache.org/luce...</a>, which my team is contributing to -- Hbase is a column-oriented DB.</p><p>We're also using Thrift quite heavily, including your earlier contribution on the asynch server stuff. I had the same reaction when I heard about Thrift: open-source protocol buffer. Did you implement a Thrift transport based on Spread? That's something we have been mulling over but haven't quite started on yet.</p><p>Chad Walters<br>Powerset</p>ChadMon, 05 Nov 2007 12:32:41 -0000