back to blog
February 21, 2014 | Jim Tommaney

InfiniDB TnT: High Speed load with InfiniDB


InfiniDB has multiple methods of loading, with different characteristics.  Loading data via Insert or “load data infile” are SQL based transactional methods, but are not suited for loading your largest tables.  InfiniDB’s fastest and most parallel load method is CPImport, the high speed data external utility.

CPImport is:

  • Scalable - each node in an InfiniDB cluster can load independently to local extents of the table.  This allows for scaling the load rate as you scale your cluster.
  • Fast - CPImport is extremely fast, loading at approximately ½ million rows per second per server, 100x to 1000x faster than the next quickest load rate.

Do you have questions about how to load your biggest tables into InfiniDB?  Let us know in the comments, below, via email at, or on Twitter @infiniDB.



jhakanchanjsr's picture



I have a table with column which would contain large text. user would need to search a word from all rows and return records which contains the word.


What could be best method to use in infinidb 4.6.

As it has mysql interface,  MySQL supports Integrated/External "Sphinx" tool. Can the same be integrated with infinidb. 


The Table has other columns ( date,number etc). the query uses various conditions along with text search.


Request your suggestions.We are in process of migrating the application to Infinidb 4.6.