SingleStore DB

Transitioning Your Application

To transition your application, simply change the connection credentials to point to SingleStore DB.

One thing to keep in mind is Code Generation. Code generation makes loading a schema into SingleStore DB for the first time is slower than with MySQL. The first time SingleStore DB encounters a table schema, it generates and compiles code that implements infrastructure around the table (memory allocation, inserts, deletes, iterations, etc.). Once a table is compiled, SingleStore DB will be able to reuse it for the lifetime of your application — even if you restart the server or drop (and recreate) the table. The INSERT queries generated by mysqldump also have to be compiled exactly once.

When transitioning your application:

  • Even if you’re connecting to SingleStore DB locally, use the explicit host 127.0.0.1 instead of localhost. Most MySQL clients will resolve localhost to use the global MySQL socket file and ignore the port setting. See the FAQ for more information.

  • As you run through the application, monitor the memsql.log file.

  • Queries that throw errors most likely correspond to unsupported syntax (see SQL Reference Overview for supported SQL surface area).