Thomas, It is quite possible that having run over it so many times in my head I was not as thurough as I should have been.
Let me clarrify. There are two possible processes. Process 1: Escalation on the ARDBC LDAP form pushes a copy of every record to form A, where all the work is performed on the data. This process causes the escalation to run on ~4500 records and takes ~7 minutes. Process 2: Escalation run on form C. This updates the one record, which causes a filter to fire that pushes an update to a display only field on the ARDBC LDAP form. This in turn causes each record to be pushed to form A, again where all the work is performed on the data. This process causes the escalation to run on one record and still process the same ~4500 records. This process takes ~12 minutes. It doesn't take a brain surgeon (which I am not) or a rocket scientist (which I am also not) to see that option one is the better way to go performance wise, however; this seems to go against what I have understood to be one of the advantages of using option 2. I hope that helps. Still, Brian Goralczyk On Thu, Mar 20, 2008 at 1:49 PM, Thomas Bean <[EMAIL PROTECTED]> wrote: > ** Brian, > Perhaps I am having a slow day, but I'm not quite clear on the scenario > you are describing. > > *"I have 4500 records on an LDAP form that need to be updated."* > Is this an ARDBC LDAP form? Are you wishing to update the original data > from your LDAP data source, or are you attempting to update/transform the > LDAP data as it is imported into another regular form in Remedy? > > *"One [option] is to use an escalation that pushes all the records, or the > other option, to use a form that initiates a push on ALL the LDAP records to > push over to the second form."* > I'm sure I'm missing something here. These two options sound more or less > identical. > With the first option, is this an escalation on the ARDBC LDAP form? > Where is the escalation pushing the records to (if not to the second form > you mentioned)? > With the second option, you mention using a "form" to initiate a push... > Is this the aforementioned "second form"? What workflow object is > initiating the push fields action (if not an escalation)? If it is a > filter, which form is the filter attached to, and what are the execution > criteria (execute on, Run If, Push Fields If)? > > --Thomas > > > ----- Original Message ----- > *From:* Brian Goralczyk <[EMAIL PROTECTED]> > *Newsgroups:* gmane.comp.crm.arsystem.general > *To:* [email protected] > *Sent:* Thursday, March 20, 2008 1:08 PM > *Subject:* Efficiency question > > ** This is more of a "looking for a consensus or opinions" type email. > > Running ARS 7.0.01 Patch 2 on a shared windows box that connected to an > Oracle database that is on a different shared box. > Using the User Tool but I don't think that matters. > > The situation: > I have 4500 records on an LDAP form that need to be updated. I am > pushing them to another form to perform the work. I have two options. One > of them is to use an escalation that pushes all the records, or the other > option, to use a form that initiates a push on ALL the LDAP records to push > over to the second form. > > I state that it is an LDAP form to indicate that I am using an external > data source, however; I don't believe that makes a difference. > > I would believe that option 2 would be quicker as it doesn't use the > single threaded escalation process, however; this is not what I have found > to be the case. There is a discrepancy of 6 minutes with the escalation > taking 7 minutes to run. As you can see, this could be quite worrisome. > > Any ideas or suggestions or even just straighten me out. > > Thanks all, > > Brian Goralczyk > __Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are" > html___ > > __Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are" > html___ _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are"
