|
I'm new to PostgreSQL, coming at it from a MySQL background. I'm currently looking at switching one of our applications (which currently uses MySQL) over to PostgreSQL and had some questions. We're considering the switch because of issues we have faced when using MySQL in a clustered scenario and we're hoping that switching to PostgreSQL will help us resolve these issues. Our three biggest pain points with MySQL are: 1. MySQL's NDB engine (used for clustering) cannot index textual data stored in a BLOB field 2. When configuring a MySQL cluster, there is a memory limit on the number of objects (tables and fields). We often have problems when importing new tables, wherein we need to increase the memory limit for the server's NDB engine before it allows us to import. This can be a problem for dynamically-generated tables, as we cannot accurately forecast the number of database objects in advance in these cases. 3. MySQL's NDB engine doesn't support or enforce foreign keys. So my first question is, I'd like to know if PostgreSQL has similar issues when running in a clustered scenario. Second, on reviewing the manual and some sites, it seems that there are a number of different OSS solutions for implementing failover and clustering with PostgreSQL, but no "official" version. Is this understanding correct? If yes, which solution is best suited for running PostgreSQL in a private cloud, with clustering/failover support? Thank you, Vikram |
