Hi,
I was interested in this:
Firebird's indexes are very dense because they compress both the prefix and
the suffix of each key. Suffix compression is simply the elimination of
trailing blanks or zeros, depending on the data type. Suffix compression is
performed on each segment of a segmented key. Prefix compression removes the
leading bytes that match the previous key. Thus a duplicate value has no key
stored at all. Dense storage in indexes minimizes the depth of the btrees,
eliminating the advantage of other index types for most data.
Do we do this? How feasible is this?
On Tuesday 16 April 2002 00:35, Christopher Kings-Lynne wrote:
> The Firebird guys have gotten around to releasing 1.0. If you read this
> front page spiel, you'll notice that they use MVCC, but with an overwriting
> storage manager.
--
Denis
