Have been thinking (and reading) a lot lately about storing data... and am about to conclude that it requires a radical shift in storing data to be able to scale (keeping cost in mind). World is talking about NoSql and its time for me to check-it-out.
I already have a system in place which works well with sql server but i wanna scale and cut the cost so thinking of either move to mysql or have mongodb along with elastic search.
These are the below options i have : -
I already have a system in place which works well with sql server but i wanna scale and cut the cost so thinking of either move to mysql or have mongodb along with elastic search.
These are the below options i have : -
Mysql with Elastic Search/Sphinx
- We can move to mysql database (which is very easy by importing).
- IDE looks fine to begin with which is workbench
- It does have INNODB to support all necessary RDBMS features and now Full text search has been added, which isnt very performant or scalable.
- We can integrate sphinx or solr with mysql, challenge is we need to convert the binary data to textual data and then only it can be indexed.
- We can also use elastic search which will be independent of database, it'll work as a rest api call.
NoSql (MongoDB) with Elastic Search
- A migration project has to be written for sql server, have to check the intricacies.
- Elastic search with river works in conjunction with MongoDB
Starting my POC on second option... Lets keep the fingers crossed.