[dba-Tech] Cassandra (was: A couple of great articles on latest and greatest)

Gustav Brock Gustav at cactus.dk
Fri May 7 12:55:27 CDT 2010


Hi Jim

It is not easy to get hold on this monster. Here are some useful links I found:

WTF is a SuperColumn? An Intro to the Cassandra Data Model:
http://arin.me/blog/wtf-is-a-supercolumn-cassandra-data-model 

Cassandra Jump Start For The Windows Developer:
http://www.coderjournal.com/2010/03/cassandra-jump-start-for-the-windows-developer/

Thrift Wiki. Basic requirements for win32:
http://wiki.apache.org/thrift/ThriftInstallationWin32

Nick Berardi's managedfusion / fluentcassandra:

FluentCassandra is a .NET library for accessing Cassandra, which wraps the Thrift client library and provides a more fluent POCO interface for accessing and querying the objects in Cassandra.
http://github.com/managedfusion/fluentcassandra/blob/master/README.mkd#readme 

Nick Berardi's C# CasandraDemo:
https://code.google.com/p/coderjournal/source/browse/trunk/Posts/2010/03/CassandraDemo.cs


So much to read and learn ...

/gustav


>>> accessd at shaw.ca 07-05-2010 16:25 >>>
For a brief moment there Terabyte was the size ultimate but that point
passed quickly and the computer world went on to Pentabyte; well that point
of ultimate has now again been surpassed as we have Zettabytes.

http://www.guardian.co.uk/technology/2010/may/03/humanity-digital-output-zet 
tabyte

Some latest information from some who have been experimenting with the new
super database called Cassandra. Notice through the performances trials the
CPU utilization remains flat! The team pushed the product to see where
better utilization could be achieved and noted setting caching would help
performance... But keep in mind this DBs performance is so far beyond our
standard SQLs.

http://jamesgolick.com/2010/4/4/two-weeks-with-cassandra.html 

An aside: It will be a while before the hard drive bottle-neck is really
fully resolved. Right now splitting a data store across numerous drives, all
indexed and cashed is the only way to lessen the performance pain. That is
why the new breed of distributive databases are so fast because they manage
the hardware layer so well. Our current crop of standard SQLs just leave the
hardware to manage it's self.
  
Jim 

_______________________________________________
dba-Tech mailing list
dba-Tech at databaseadvisors.com 
http://databaseadvisors.com/mailman/listinfo/dba-tech 
Website: http://www.databaseadvisors.com





More information about the dba-Tech mailing list