Jim Lawrence
accessd at shaw.ca
Thu Feb 21 12:07:01 CST 2013
Hi Shamil: A very interesting subject. There are so many questions about how to properly manage data when data is huge and the number of users are massive. This lends its self to the NoSQL world of huge data where data can be spread across thousands of servers being accessed and changed and added to by thousands of users. I have noticed this latency with Facebook where just running a refresh may display one set of data then a new refresh and another slightly different set is presented...and the data isn't new...but all the matching results are not retrieved at any single time. Multiple refreshes, in a row, seems to produce the best results. So what types of data and database will you use. It will be difficult to emulate the same set of issues when a system only has limited users and data...doesn't such tests require a high stressed environment? But do keep us posted on the scenario and products you will be using to test these concepts and any significant results you subsequently observe. Jim -----Original Message----- From: dba-tech-bounces at databaseadvisors.com [mailto:dba-tech-bounces at databaseadvisors.com] On Behalf Of Salakhetdinov Shamil Sent: Wednesday, February 20, 2013 4:57 PM To: Discussion of Hardware and Software issues Subject: [dba-Tech] CQRS or 6*9 = 42 Hi All -- I'm considering to try to apply CQRS + ES ( Command & Query Responsibility Segregation and Event Sourcing) architectural patterns http://www.udidahan.com/2009/12/09/clarified-cqrs/ http://www.udidahan.com/2010/05/07/cqrs-isnt-the-answer-its-just-one-of-the- questions/ in one of my test projects. I'm still to find good practical info on CQRS + ES drawbacks - do you have some? Thank you. -- Shamil _______________________________________________ dba-Tech mailing list dba-Tech at databaseadvisors.com http://databaseadvisors.com/mailman/listinfo/dba-tech Website: http://www.databaseadvisors.com