Re: Performance Monitoring/Notification/Reporting

[rob@xxxxxxxxx]

Very well thought and modernized question.

You and I have similar workloads I believe.  My boss, (whom you've met), 
is using iSeries Navigator to do must of his performance monitoring.

We've not yet tied the monitors into Sametime, but we're interested in 
doing so.  By end of week I will have Sametime Everyplace installed so 
that I can use my Treo 600 to do Sametime.  Be nice to have this for 
monitoring.

Response time metrics in a "non-traditional" environment.  This is where 
our concern is.  For example, when only 9% of your system is used for 
"traditional" applications like payroll, ERP, accounting, what is good, 
bad, meaningless for page faulting in *base?  However, you have it right 
when you state that the end result you want is acceptable response time. 
Looking at page faulting is a possible symptom - not the problem.

Is using the same performance arguments from an handout prepared for 
COMMON in 1988 still valid?  For example, I know that 7 newer disk drives 
with newer cache, etc blow the doors off of 42 older disk drives.  Rather 
destroys the disk arm mantra.

We have some homegrown applications for monitoring the status of our 
Domino applications and performing alerts.  But are interested in 
exploring Domino Administrator capabilities more to see what's built in 
there.  For example, our sametime domino servers start, but often the 
sametime services themselves try to start and then peter out.  Our 
existing homegrown doesn't go to this granularity.  I need to.  Some of 
our homegrown is more systemic.  For example, we send out an email from 
Domino to an external service and get a response.  If the response doesn't 
come back in a specified time, then action needs to be taken.  The problem 
may be with:  Domino, i5, network, external service, inbound only, 
outbound only, etc.  We get notified when it doesn't work, but we need to 
analyze why.  We have this process constantly running.

Speaking of the Sametime Everyplace, I've had two people pop in just 
during the typing of this email wanting to know the status - got to go.

Rob Berendt
-- 
Group Dekko Services, LLC
Dept 01.073
PO Box 2000
Dock 108
6928N 400E
Kendallville, IN 46755
http://www.dekko.com





Mike.Crump@xxxxxxxxxxxxxxxx 
Sent by: midrange-l-bounces@xxxxxxxxxxxx
01/05/2005 09:48 AM
Please respond to
Midrange Systems Technical Discussion <midrange-l@xxxxxxxxxxxx>


To
midrange-l@xxxxxxxxxxxx
cc

Subject
Performance Monitoring/Notification/Reporting






Broad question but I was looking for some insight on what people were 
doing
for perf. monitoring and reporting (including ISV, PT/400, iSeries
Navigator monitors, etc.).  Today, I'm working mostly with PT/400 and
iSeries Navigator but I want to broaden my abilities.  I think I can do it
with these products but that some functions will require some effort.  Not
that I'm against it but I just don't have the time I used to have..... :-)

My broad requirements are:

1.)  The usual metrics (cpu utilization, interactive utilization, database
utilization, interactive response time, transaction rate, disk arm
utilization, dasd iop utilization, dasd utilization, memory pool fault
rates, logical IO, physical IO, etc.)  Which does beg a question - with
WAS, Domino, HTTP, QP, ST, ODBC, etc. what are people's thoughts on
response time metrics?  We keep on adding more and more non-green screen
applications so my interactive response time is now only a portion of my
response time concerns.

2.)  Real time monitoring with real time viewing.  Close to real time, 
none
of this wait till tomorrow's reports.

3.)  Drill down capabilities from online monitors - ability to drill down
to a job.

4.)  Alert processing - ability to set thresholds and multiple alert
capabilities - message queue, pager, email, SameTime, etc.  Alert 
reporting
- summary of alerts for a given time frame.

5.)  Executive report capabilities through the web.

6.)  Very granular control over what get's monitored and reported on.

7.)  Ability to granularly control historical measures.  In essence, roll
real time data to a daily summary, to a monthly summary, to a yearly
summary, etc.  And save DASD storage while I'm at it.

8.)  Various publishing capabilities - prefer presenting data for the
browser but may need to also automatically publish to PDF or some other
type of output.

9.)  Plug-in's for Domino, WAS, etc.  I see a need to present system data
but may also need to start looking more closely at providing some ability
to monitor/report for application servers......



Michael Crump
Manager, Computing Services
Saint-Gobain Containers
1509 S. Macedonia Ave.
Muncie, IN  47302
(765)741-7696
(765)741-7012 f
(800)428-8642

"The probability that we may fail in the struggle ought not to deter us
from the support of a cause we believe to be just"  Abraham Lincoln




-- 
This is the Midrange Systems Technical Discussion (MIDRANGE-L) mailing 
list
To post a message email: MIDRANGE-L@xxxxxxxxxxxx
To subscribe, unsubscribe, or change list options,
visit: http://lists.midrange.com/mailman/listinfo/midrange-l
or email: MIDRANGE-L-request@xxxxxxxxxxxx
Before posting, please take a moment to review the archives
at http://archive.midrange.com/midrange-l.