|
|
|
c-treeACE V10 represents a major engineering milestone related to Partition File support. The ability to define criteria rules that direct data I/O to separate physical data files (partitions) is a powerful feature.
An extensive focus on the speed and performance of partitioned files highlighted numerous areas for increased efficiency and performance gains. Improvements in internal indexing strategies resulted in significant performance overall. A number of other issues were also addressed.
Global unique indices, range requests, alter table support, Windows/Unix time-stamp support, and partition administration have all been greatly improved. These enhancements are most evident when using c-treeACE SQL over a partitioned file.
For those new to c-treeACE, a partitioned file logically appears to be one file (or more accurately, one data file and its associated index files); however, it is actually comprised of a set of files whose contents are partitioned by the partition key. Both data and index files are partitioned. This permits data with a defined range of values for the partition key to be rapidly purged or archived rather than a record-by-record delete within this range. For example, a single date range can be dropped with a single call.
For more information, see Partitioned Files.
