How to speed up the memory database through special data type index

How to speed up the memory database by indexing special data types

Let's take a look at the full-text index （Full-Text Index）. Full text search uses reverse indexing to search for words or phrases in a large number of texts . This kind of search can be precise or fuzzy , Currently only supported CHAR、VARCHAR、TEXT and LONGTEXT These data types . For now RapidsDB Come on , Full text indexing is only supported on columnar tables . Besides , Full-text index CREATE TABLE Part of the query is enabled . This means that the full-text index cannot be deleted or changed after the table is created . If the table is deleted , Then the index will be automatically deleted .

Take another look at geospatial index . Geospatial indexes are rarely used in domestic projects , It is stored in the column where the spatial data is stored , Used to define geospatial indexes , And used to speed up their query . Geospatial indexes use control parameter values （6-32） Segment polygons and linetypes . The smaller the number, the smaller the memory consumption of the index 、 The faster operations such as inserting and updating , But the query time is slow . The larger the control parameter value, the higher the query performance at the cost of memory and insertion performance .

In this way, we can analyze the business application , Adjust the index key in response , Get the optimization effect of improving speed . Okay , Let's introduce it here .RapidsDB Extreme performance database is still worthy of further attention .