A few questions

Hello All, A few of questions: --- In 7.3 I see that after backing up old log files, they get marked U and B. In 5.0, I see that instead of marking the logs U (used), and B (backed-up), it marks them as F (free). Is this behavior normal, or is something wrong with my 7.3 instance. I am just wondering when those logs marked as U and B become F (free). --- There is no USERS parameter in the onconfig file for 7.3. Does this mean that 7.3 has hardcoded limits for the # of users? --- I read a few posts a while back about the possibility of the optimizer being too slow to determine a query path. I believe it was on a 7.3X instance. My main question is, I myself have dissapointed by the raw speed of 7.3 vs 5.0, and is this a likely place to look for possible performance improvements? If so, is there anyway within 4gl RDS to code apps to minimize the length of time the optimizer evaluates its query plan possibilities? --- If I want to split high use tables across drives, what are some things that I should consider when deciding what goes where. Thanks in advance. Olaf

rico@wsx.wsex.com wrote in message <7fo1rt$8mq$1@news.xmission.com>... > >Hello All, > >A few of questions: > >--- >In 7.3 I see that after backing up old log files, they get marked U and B. >In 5.0, I see that instead of marking the logs U (used), and B (backed-up), it marks them as F (free). >Is this behavior normal, or is something wrong with my 7.3 instance. >I am just wondering when those logs marked as U and B become F (free). > Everything is normal in 7.3..... They just changed the notation. In certain cases (after a restore, for example) the logs will be marked free (F) again. but U-- and U-B are the norms now. >--- >There is no USERS parameter in the onconfig file for 7.3. >Does this mean that 7.3 has hardcoded limits for the # of users? > >--- >I read a few posts a while back about the possibility of the optimizer being too slow to determine a query path. >I believe it was on a 7.3X instance. >My main question is, I myself have dissapointed by the raw speed of 7.3 vs 5.0, and is this a likely place to look for possible performance improvements? >If so, is there anyway within 4gl RDS to code apps to minimize the length of time the optimizer evaluates its query plan possibilities? > is your OPTCOMPIND in onconfig file set to 0?? IF not, it should be. >--- >If I want to split high use tables across drives, what are some things that I should consider when deciding what goes where. > simplest approach would be to 'round robin' the fragments of a table. And, obviously, put chunks on separate drives (channels, if you can) >Thanks in advance. > >Olaf Just my opinions, for what they're worth ($0.02 gratefully accepted as generous) Doug Agnew DBA Charlotte Pipe & Foundry dagnew@charlottepipe.com

rico@wsx.wsex.com wrote: > > Hello All, > > A few of questions: > > --- > In 7.3 I see that after backing up old log files, they get marked U and B. > In 5.0, I see that instead of marking the logs U (used), and B (backed-up), it marks them as F (free). > Is this behavior normal, or is something wrong with my 7.3 instance. > I am just wondering when those logs marked as U and B become F (free). Yes all is OK. IDS 7.xx does mark the logs free but only for a fraction of a second immediately before reusing them so you never see it unless you are lucky (Newly added log files are also free and there is also some situation where the logs are marked free when the engine first starts up but I've never figured out when that is exactly). This was done so the logs that have been backed up will still be viewable with onlog. Onlog will not look at a 'free' log assuming that is has just been added and is therefore empty. > There is no USERS parameter in the onconfig file for 7.3. > Does this mean that 7.3 has hardcoded limits for the # of users? No but the equivalent memory structure that the USERS parameter used to size at startup is now dynamic so that parameter is no longer required. The numbber of active sessions is now controlled by the NETTYPE parameter mostly since that allocates the number of listener data structures used for concurrent sessions. > I read a few posts a while back about the possibility of the optimizer being too slow to determine a query path. > I believe it was on a 7.3X instance. See my answer to a post today from Russ Evans: "OPTCOMPIND = 0: Why is this always faster?" for one cause of slow query path calculation. > My main question is, I myself have dissapointed by the raw speed of 7.3 vs 5.0, and is this a likely place to look for possible performance improvements? > If so, is there anyway within 4gl RDS to code apps to minimize the length of time the optimizer evaluates its query plan possibilities? SET OPTIMIZATION LOW; > If I want to split high use tables across drives, what are some things that I should consider when deciding what goes where. Table interaction, position of the chunk's partition (or offset) on the drive (see the Performance Guide for notes about track positioning), the individual capabilities of each drive (if one is a 7200 RPM spindle it will be slower to return data than one with a 10000 RPM spindle speed. Art S. Kagel