As of 1.10-beta, RT indexes are a beta quality feature: while no major,
showstopper-class issues are known, there still are a few known usage quirks.
Those quirks are listed in this section.
- Prefix and infix indexing are not supported yet.
- MVAs are not supported yet.
- Disk chunks optimization routine is not implemented yet.
- On initial index creation, attributes are reordered by type,
in the following order: uint, bigint, float, timestamp, string. So when
using INSERT without an explicit column names list, specify all uint
column values first, then bigint, etc.
- Default conservative RAM chunk limit (
rt_mem_limit)
of 32M can lead to poor performance on bigger indexes, you should raise it to
256..1024M if you're planning to index gigabytes. - High DELETE/REPLACE rate can lead to kill-list fragmentation
and impact searching performance.
- No transaction size limits are currently imposed;
too many concurrent INSERT/REPLACE transactions might therefore
consume a lot of RAM.
- In case of a damaged binlog, recovery will stop on the
first damaged transaction, even though it's technically possible
to keep looking further for subsequent undamaged transactions, and
recover those. This mid-file damage case (due to flaky HDD/CDD/tape?)
is supposed to be extremely rare, though.
- Multiple INSERTs grouped in a single transaction perform
better than equivalent single-row transactions and are recommended for
batch loading of data.
