Re: How does WebCatalog search the database?
This WebDNA talk-list message is from 1997
It keeps the original formatting.
numero = 15483
interpreted = N
texte = >But my question is: How does WebCatalog search its data? >As it uses a flat file databases one should expect it to do a search >through the entire file before returning the matching data.That's right!>This is, as I understand it, not a good way if the database >contains _many_ records.That *may be* right when the data must be retrieved from a hard drive, but it's very *WRONG* when the data is cached in RAM -- and searched in RAM -- the way WebCatalog does it.>So my question is: does WebCatalog index the data when it loads >a database into memory?No, and it really doesn't have to. You're still thinking in terms of searching for data when it resides on a hard drive instead of in RAM. This slow searching problem you're imagining has nothing to do with WebCatalog, because WebCat caches its entire database in RAM.That means no indexes are ever needed, and it also means a text search thru the whole database is MUCH FASTER than any of the disk-based searches you're probably used to.In other words, none of the data in any WebCatalog database file has to be indexed, because it's already in RAM, ready for searching at the speed of electrons. WebCatalog's technique totally avoids the slowdowns inherent in waiting for a mechanical hard drive to locate the proper data on disk ... so it far exceeds anything that can be accomplished by indexing and searching a hard drive.>If not, how does WebCatalog perform when it comes to BIG databases?How big is BIG? Some WebCatalog users have between 200,000 and 300,000 records in their databases, and it's still far faster than any of the alternatives ... :)Sincerely, Ken Grome ken@iav.com WebDNA Solutions http://www.hui.net/
Associated Messages, from the most recent to the oldest:
>But my question is: How does WebCatalog search its data? >As it uses a flat file databases one should expect it to do a search >through the entire file before returning the matching data.That's right!>This is, as I understand it, not a good way if the database >contains _many_ records.That *may be* right when the data must be retrieved from a hard drive, but it's very *WRONG* when the data is cached in RAM -- and searched in RAM -- the way WebCatalog does it.>So my question is: does WebCatalog index the data when it loads >a database into memory?No, and it really doesn't have to. You're still thinking in terms of searching for data when it resides on a hard drive instead of in RAM. This slow searching problem you're imagining has nothing to do with WebCatalog, because WebCat caches its entire database in RAM.That means no indexes are ever needed, and it also means a text search thru the whole database is MUCH FASTER than any of the disk-based searches you're probably used to.In other words, none of the data in any WebCatalog database file has to be indexed, because it's already in RAM, ready for searching at the speed of electrons. WebCatalog's technique totally avoids the slowdowns inherent in waiting for a mechanical hard drive to locate the proper data on disk ... so it far exceeds anything that can be accomplished by indexing and searching a hard drive.>If not, how does WebCatalog perform when it comes to BIG databases?How big is BIG? Some WebCatalog users have between 200,000 and 300,000 records in their databases, and it's still far faster than any of the alternatives ... :)Sincerely, Ken Grome ken@iav.com WebDNA Solutions http://www.hui.net/
Kenneth Grome
DOWNLOAD WEBDNA NOW!
Top Articles:
Talk List
The WebDNA community talk-list is the best place to get some help: several hundred extremely proficient programmers with an excellent knowledge of WebDNA and an excellent spirit will deliver all the tips and tricks you can imagine...
Related Readings:
Plugin or CGI or both (1997)
Single Link browsing (1997)
(2003)
Mime Headers for Mail (1998)
Re:Emailer setup (1997)
RE: pop up building with summarize (1998)
PhotoMill -> PhotoMaster (1997)
[WebDNA] function variables (2011)
Converting a Magazine to a Web Site with WebCat (1999)
ShipCost Data Base (1998)
SM: Grep Syntax? (2005)
What am I missing (1997)
Netscape 3.01 can't see db in form (1997)
Sorting problem (1997)
Unusual characters in the db ... (2002)
[WebDNA] Use of MySQL with WebDNA (2010)
Dates - Search and Sorting (1999)
[ShowNext] feature in 2.0 (1997)
Robert Minor duplicate mail (1997)
Errant Email. (1998)