![]() | ![]() Products ![]() ![]() ![]() ![]() |
Home » Technical Support » ElevateDB Technical Support » Support Forums » ElevateDB General » View Thread |
Messages 111 to 120 of 146 total |
![]() |
Tue, Oct 3 2006 3:54 PM | Permanent Link |
Tim Young [Elevate Software] Elevate Software, Inc. ![]() | Michael,
<< Both Table and Query. >> Actually it's quicker just to simply count up to the current position. The maintenance alone on such an index is very time-consuming since it has to be constantly synched to the current state of the table by rebuilding portions of it. -- Tim Young Elevate Software www.elevatesoft.com |
Tue, Oct 3 2006 3:59 PM | Permanent Link |
Tim Young [Elevate Software] Elevate Software, Inc. ![]() | AJ,
<< Ok why not give a list of future features that will be effected and how bad. Then people can decide if 3 state scrollbars are really that bad. >> I've already mentioned a big one - having versioning of the index keys for multi-versioning in the transactions. Also, more esoteric locking in the indexes for greater concurrency. When the entire path from the root to the leaf of the index tree needs updated for every insertion or deletion into the tree, then the entire path must always be locked. Finally, fail-safe writes need to log all I/O before actually applying it to the table, thus any decrease in the amount of I/O makes fail-safe writes more and more doable. And fail-safe writes mean, barring complete hardware failure, no more corruption or repairing. -- Tim Young Elevate Software www.elevatesoft.com |
Tue, Oct 3 2006 4:01 PM | Permanent Link |
Tim Young [Elevate Software] Elevate Software, Inc. ![]() | Stefano,
<< I am waiting for the first release.....(I vote for remove stat). >> Thanks. << Excuse me for my elementary concepts, but my english is so bad that I'm not sure that you understand me, sorry ![]() I think I understood what you said - you were asking whether the record x of x type of display would still be possible when using TEDBTable components. I replied that it isn't impossible yet, but it might be if we remove the index stats like we're discussing right now. -- Tim Young Elevate Software www.elevatesoft.com |
Tue, Oct 3 2006 9:26 PM | Permanent Link |
Steve Gill | Hi Tim,
> Finally, fail-safe writes need to log all I/O before actually applying > it to the table, thus any decrease in the amount of I/O makes fail-safe > writes more and more doable. And fail-safe writes mean, barring > complete hardware failure, no more corruption or repairing. Well, to me that sounds like a great reason to dump the stats. ![]() Regards, SteveG |
Wed, Oct 4 2006 2:57 AM | Permanent Link |
Chris Holland SEC Solutions Ltd. ![]() | Given the option I would choose fail-safe writes over stats. Chris Holland Tim Young [Elevate Software] wrote: > AJ, > > << Ok why not give a list of future features that will be effected and how > bad. Then people can decide if 3 state scrollbars are really that bad. >> > > I've already mentioned a big one - having versioning of the index keys for > multi-versioning in the transactions. Also, more esoteric locking in the > indexes for greater concurrency. When the entire path from the root to the > leaf of the index tree needs updated for every insertion or deletion into > the tree, then the entire path must always be locked. Finally, fail-safe > writes need to log all I/O before actually applying it to the table, thus > any decrease in the amount of I/O makes fail-safe writes more and more > doable. And fail-safe writes mean, barring complete hardware failure, no > more corruption or repairing. > |
Wed, Oct 4 2006 3:40 AM | Permanent Link |
Roy Lambert NLH Associates ![]() | If the fail-safe applies to fileserver mode as well then no stats becomes a no brainer (DBISAM is good but I do have to do repairs occasionally)
Roy Lambert |
Wed, Oct 4 2006 11:01 AM | Permanent Link |
"Jerry Hayes" | > Given the option I would choose fail-safe writes over stats.
This decision keeps getting easier and easier ![]() |
Wed, Oct 4 2006 3:30 PM | Permanent Link |
"R. Tipton" | Tim,
"Tim Young [Elevate Software]" <timyoung@elevatesoft.com> wrote in message news:20C40701-E4BF-4F91-BA73-986B900695F3@news.elevatesoft.com... > > The effect of the index statistics on DISTINCT operations is minimal, > unfortunately. The only thing that DBISAM and ElevateDB don't do with > regard to distinct processing in terms of nice "shortcuts" is use existing > indexes, which is what I believe SFD does. > Well you asked for feedback and you got it, most peeps say dump the stats. Rita |
Wed, Oct 4 2006 3:47 PM | Permanent Link |
"B Miller" | I agree. I will take performance and stability over stats. I use DevEx
Grids so the scroll bar is not an issue. Bill |
Wed, Oct 4 2006 4:12 PM | Permanent Link |
Heiko Knuettel | I vote for working scrollbars...it would be a nightmare to replace every
descended DBGrid in my apps |
« Previous Page | Page 12 of 15 | Next Page » |
Jump to Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
This web page was last updated on Saturday, June 22, 2024 at 05:51 PM | Privacy Policy![]() © 2024 Elevate Software, Inc. All Rights Reserved Questions or comments ? ![]() |