Choosing a Database Page Size

From InterBase

Go Up to Database Properties

InterBase pages are 4KB by default. A typical production InterBase database gains 25 to 30 percent performance benefit by using this page size, relative to smaller page sizes. This page size results in better performance for the following reasons:

  • Fewer record fragments are split across pages
It is common for records to be larger than a single page. This means that InterBase fragments records and stores them on multiple pages. Querying a given record requires multiple page reads from the database.
By increasing the size of a page, InterBase can reduce the number of multiple page reads and can store record fragments more contiguously.
  • Index B-trees are more shallow
Indexes are B-trees of pointers to data pages containing instances of specific indexed values. If the index B-tree is larger than one page, InterBase allocates additional database pages for the index tree. If the index pages are larger, InterBase needs fewer additional pages to store the pointers. It is easier for the database cache to store the entire B-tree in memory, and indexed lookups are much faster.
  • I/O is more contiguous
It is fairly likely for a query to request successive records in a table. For example, this is done during a table scan, or query that returns or aggregates all records in a table. InterBase stores records on the first page that is unused, rather than ensuring that they are stored near each other in the file. Doing a table scan can potentially require retrieval of data by seeking all over the database. Seeks take time just as reading data takes time.
Any given page can store records from only one table. This indicates that a larger page is certain to contain more data from the same table, and therefore reading that page returns more relevant data.
  • Default number of cache buffers is a larger amount of memory
InterBase allocates the database cache in number of pages, rather than a fixed number of bytes. Therefore defining a larger page size increases the cache size. A larger cache is more likely to have a better hit rate than a smaller cache.
  • Most operating systems perform low-level I/O in 4096 byte blocks
InterBase performs a page read or write at the OS level by reading in 4096 byte increments regardless of the size of the database page. Therefore, by defining the database with a page size of 4096, the database I/O matches the low-level I/O and this results in greater efficiency when reading and writing pages.

Although 4KB seems to be the best page size for most databases, the optimal size depends on the structure of the specific metadata and the way in which applications access the data. For this reason, you should not consider the 4KB page size guideline to be a magic value. Instead, you should perform testing with your application and database under several different page sizes to analyze which configuration gives the best performance.

Advance To: