YetAnotherForum
Welcome Guest Search | Active Topics | Log In | Register

Totalizer Subsystem Options · View
ThePIman
#1 Posted : Thursday, May 14, 2009 12:39:23 PM
Rank: Advanced Member
Groups: Member

Joined: 5/14/2009
Posts: 35
Location: UK
Hi,

We are not currently using any Totalizer tags but it has been suggested we move some of our daily calcs to them.
As I'm sure you already know, time series PE functions like TagTot can be archive intensive when looking for the last day of data (288 * 5 minute readings), and when you have thousands of calc tags doing this they could make other scheduled scans/calcs skip.

I was advised to put Totalizers in for our PEs that use built in functions like TagTot, TagAvg, findeq, findXX, timeeq, timeXX2 etc., but I have reservations about the Totalizer system eg:
- max 5000 events
- values not stored (ie no recalculations etc)

PI Ace is not currently available, so I would prefer to use a cumulative total PE instead that runs every 5 minutes and resets at midnight:
IF yearday(Prevevent('sourcetag', '*-5m')) = yearday('y') THEN 'sourcetag' ELSE ('ThisTag' + 'sourcetag')

But what are other peoples views on the Totalizer? Does it cause more problems than it is worth ?

Thanks


.
Sponsor  
 

OSIsoft vCampus is a subscription-based, online offering that consists of providing everything people need to develop applications on the PI System.
We invite you to take a "tour" of the OSIsoft Virtual Campus - also feel free to consult the FAQ  or contact OSIsoft vCampus for more details.
Pho
#2 Posted : Thursday, January 14, 2010 2:35:06 PM
Rank: Member
Groups: Member

Joined: 1/14/2010
Posts: 28
Location: UK
Hi ThePiMan,

We are in the same boat.
We now have that many tags, the tagtots are causing the 5 min calcs to be pushed back.

We were advised to use totalizers as well, but like you said, the idea of these being unable to recalculate is a big issue for us.
They are fine when everything is ok and running,but if there is a down period, they dont work.

to get round the problem of tag tots, i have basically added more scan rates around the midnight hour, with longer off sets and change the calc to be along the lines of:

TagTot('sourcetag', 'Y', 'T')

that way, by using the Y and the T, is doesn't matter when the calc runs, it picks up the time stamps from Yesterday @ midnight and Today @ midnight.
it seems to have helped quite a bit.
Michael
#3 Posted : Thursday, January 14, 2010 4:28:28 PM
Rank: Advanced Member
Groups: Member

Joined: 12/3/2009
Posts: 71
Location: Germany/Pennsylvania
Hi,

We also did such calculations using PE tags executed by dedicated scan classes, exactly for the same reasons described above. In addition there is one subsystem less to run :-)

Depending on your compression settings the quality of the data is not better.
ThePIman
#4 Posted : Friday, January 15, 2010 8:22:24 AM
Rank: Advanced Member
Groups: Member

Joined: 5/14/2009
Posts: 35
Location: UK
Hey Pho,

Check the date of my post.
Your boat WAS my boat at that time.
Your solution sounds very familiar, but I never had the time to implement it d'oh!

Welcome to the boards

.
Users browsing this topic
Guest
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.