Clever Geek Handbook

How does Sphinx store null fields and how does this affect performance?

I am thinking of using sphinx to index many fields (in hundreds), many of which will be null. My question is, how much will the number of fields affect performance?

This situation does not arise from incredibly denormalized data, but from the requirements for the search interface and what to look for. Basically I will be dynamically building an index configuration in an indexed model and may end up with quite a lot of null fields.

My guess is the performance / success of this depends on what Sphinx does with null values ​​... if it just ignores them then I should be fine, but if it actually keeps the field at index zero in its index I can there is a problem.

0


source to share


2 answers


Apparently the current version of Sphinx is limited to 32 text fields per index (unlimited numeric), so I would have to split the data into multiple different indices to achieve this if all the data could not be represented numerically (which may not be mine). Go back to the drawing board!



0


source


A large number of null fields will not greatly impact performance. The main problem with many null fields is that they will eat a good chunk of memory.



0


source






More articles:


All Articles