Re: access path

[Vern Hamberg <vhamberg@xxxxxxxxxxxxxxxxxxxxxxxxx>]

Leif

I've always heard it described as a description of how to have access to 
the records of a file. This can be sequential or keyed, based on RRN 
("keyed" on RRN?) - of course, RRN can change, a key should not - or on a 
key value.

It's not just an iSeries term. DB2 for various platforms also uses it. A 
brief google reveals the equivalency of "access path" and "index"

> Access path of file was used by query. Indicates that an existing
> index was chosen to process the query. (iSeries)

> The graph of the access path is displayed on an IBM OS/2® or Microsoft® 
> Windows
> NT® workstation, eliminating the need to interpret the plan table output 
> (OS/390 DB2 Visual Explain)

An unkeyed physical file (probably an iSeries-only concept) can be said to 
have an arrival sequence access path.

Rebuilding? Depends on what you mean. An index is maintained every time a 
record is added/removed from the physical file. The tree may need balancing 
from time to time. A complete rebuild is probably rare on the 400.

Regards

Vern

At 10:05 PM 4/23/2003 -0500, you wrote:
> I have been working on and with this machine for upwards
> of two sunspot cycles, and I have a confession to make:
> I see references on this list and elsewhere to an "access
> path" to a file. I have studied the internals of this machine,
> I have looked far and wide, but I'm still baffled: what the
> h*** is an "access path"? where is it? what does it do?
> Why must it be "rebuilt" at times and what does that entail?
> Somebody enlighten me...