The Association of System

Performance Professionals

The Computer Measurement Group, commonly called CMG, is a not for profit, worldwide organization of data processing professionals committed to the
measurement and management of computer systems. CMG members are primarily concerned with performance evaluation of existing systems to maximize
performance (eg. response time, throughput, etc.) and with capacity management where planned enhancements to existing systems or the design of new
systems are evaluated to find the necessary resources required to provide adequate performance at a reasonable cost.

This paper was originally published in the Proceedings of the Computer Measurement Group’s 2005 International Conference.

For more information on CMG please visit http://www.cmg.org

Copyright Notice and License

Copyright 2005 by The Computer Measurement Group, Inc. All Rights Reserved. Published by The Computer Measurement Group, Inc. (CMG), a non-profit
Illinois membership corporation. Permission to reprint in whole or in any part may be granted for educational and scientific purposes upon written application to
the Editor, CMG Headquarters, 151 Fries Mill Road, Suite 104, Turnersville , NJ 08012.

BY DOWNLOADING THIS PUBLICATION, YOU ACKNOWLEDGE THAT YOU HAVE READ, UNDERSTOOD AND AGREE TO BE BOUND BY THE
FOLLOWING TERMS AND CONDITIONS:

License: CMG hereby grants you a nonexclusive, nontransferable right to download this publication from the CMG Web site for personal use on a single
computer owned, leased or otherwise controlled by you. In the event that the computer becomes dysfunctional, such that you are unable to access the
publication, you may transfer the publication to another single computer, provided that it is removed from the computer from which it is transferred and its use
on the replacement computer otherwise complies with the terms of this Copyright Notice and License.

Concurrent use on two or more computers or on a network is not allowed.

Copyright: No part of this publication or electronic file may be reproduced or transmitted in any form to anyone else, including transmittal by e-mail, by file
transfer protocol (FTP), or by being made part of a network-accessible system, without the prior written permission of CMG. You may not merge, adapt,
translate, modify, rent, lease, sell, sublicense, assign or otherwise transfer the publication, or remove any proprietary notice or label appearing on the
publication.

Disclaimer; Limitation of Liability: The ideas and concepts set forth in this publication are solely those of the respective authors, and not of CMG, and CMG
does not endorse, approve, guarantee or otherwise certify any such ideas or concepts in any application or usage. CMG assumes no responsibility or liability
in connection with the use or misuse of the publication or electronic file. CMG makes no warranty or representation that the electronic file will be free from
errors, viruses, worms or other elements or codes that manifest contaminating or destructive properties, and it expressly disclaims liability arising from such
errors, elements or codes.

General: CMG reserves the right to terminate this Agreement immediately upon discovery of violation of any of its terms.
 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

CAPTURING MID-RANGE SYSTEMS DATA USING NATIVE COMMANDS

Robert F Patterson – Bank of America

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Our company was paying large licensing fees for third-party products for gathering
systems data for capacity planning and performance management. The focus of
our effort was to save cost and at the same time obtain data for more mid-range
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

systems (Unix, Linux, and Windows servers). We wrote scripts to gather our own
data using native commands.

This paper compares the various native commands and metrics capturing data in
AIX, Solaris, HP-UX, and Linux, and presents an overview of the capture
methodology. It also presents a high-level overview of how the data is used.

Windows data is captured in a similar manner, but is not covered in this paper.

Introduction limits to the requirements definition process, basically

taking vital sign measurements that also happen to be
We were using third-party products to capture systems good for both capacity planning and performance
data from mid-range systems (Unix, Linux, and tuning purposes. Our philosophy has been that when
Windows servers) to use for capacity planning and the server is healthy there is no compelling reason to
resolving performance issues. We had invested collect more data, and when problems are
substantial time and effort in creating our own scripts encountered, a more in-depth analysis is performed.
to manage and present the collected data. However, The results of this initial requirements definition effort
we were dissatisfied for a number of reasons: are documented in Table A.

 Licensing costs were high. Defining Native Commands

 Because of contract and budgeting limits, we had
to pick and choose where to install; we could not We next researched the native commands which
install everywhere. would acquire the required data elements. It was
 We were unable to negotiate enterprise licenses. apparent that different operating systems were very
 Implementing new collectors could be laborious; to different for some of the individual data elements and
simplify, we had to initiate 24-hour collection each quite a bit of research was required for some of them.
day. For example, “prtdiag” is a Sun-only command. Even
 Data presentation capabilities were not SAR data differs among operating systems. Table B
acceptable; we usually used other utilities, e.g. reflects a subset of the native commands we
Excel, to generate desired graphs. researched, but it illustrates the variety among
operating systems.
Another unit within the company had developed scripts
utilizing native operating system commands to gather
system data. We decided to expand their approach to Building Collectors
accommodate our needs.
Using defined native commands, we then built
collectors using a common script template, but
Requirements accounted for differences in operating systems. The
general features of our collectors are:
Our first step was to create a set of requirements for
the data needed for Unix servers. We also requested • Installation packaging is provided by our Sysadmin
reporting requirements. We gathered input from all of groups.
the units which would be utilizing the collected data. • Collector installation creates collect, archive, and
Generally our experience was that the “wish” lists scripts directories, as well as creating cron entries
became too cumbersome, generally translating into to initiate 5-minute and daily executions (also
“everything”. We had to set some boundaries and implying that collection begins immediately upon

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

installation). necessary, e.g. load testing exercises.

− The collect directory contains today’s currently • Users locate and extract required information,
collected data (see Table C). downloading the selected data to their
− The archive directory contains tar’ed and workstations.

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
zipped files containing all of the detail data for • Data can be accessed for multi-server and multi-
a single day (see Table C). day reporting.
• SA data in the archive directory is converted to • Utilities are provided for converting the data to .csv
SAR reports at end of day. format.
• Configuration information is captured at end of
day. Since the data is now in .csv format, graphs can be
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

• The archive files are retained on the agent servers created using Excel or any other analysis software that
for 7 days, and then removed. This means that 7 accepts .csv files.
full days of data are retained on the agent server,
even when data is transferred to the datahub. Excel is broadly used for 3 reasons:

Table D shows the actual native commands we are • It is universally available on every laptop.
using. In some cases, minor editing of data occurs on • We don’t have to educate the end user.
the agent server as the data is collected (no specific • The data resides in flat files on multiple hubs and
command output is presented). must be copied down to the user laptop prior to
analysis.

Managing Data If users want to use some other approach after they
copy it to their workstation (such as SAS or
We established six Windows datahubs to collect and GNUPLOT) they can, but we don’t provide support with
store our data. Activities on the datahubs include: the staff we have. Most other tools require the data to
be in a database on the hubs; we chose not to provide
• Data is transferred to the datahubs on a daily that type of support.
basis.
• Transfer is controlled by limiting transfers to four at Using the Data – Capacity Planning
a time; normally data transfer is finished shortly
after midnight. Most of our capacity planning analysts use Excel to
• For any given agent server, scripts on the datahub generate their analysis charts. CPU utilization is the
check what is already on the datahub and only variable most often used.
transfer files that are needed from the agent
server (remember that the agent server keeps We use Specints for comparison of computer speed
data for the last 7 days). (integer operation/sec measure of raw CPU speed)
• The data initially is transferred to a working area across different models and even vendors. We also
where it is “massaged”: frequently use transactions/second parameters that
− White space is compressed. are defined empirically for specific hardware and
− Inactive disk data records are removed from applications by our testing labs.
SAR reports.
− Zero length files are removed. Examples of Excel generated capacity planning charts
are reflected in Diagram E.
• Data is then moved to permanent directories.
• Data will be removed after one year.
• Granularity is maintained for the entire life of the
data; no peak/valley summarizations are
performed. Using the Data – Dashboards

The web browser has become the standard for

Accessing Data delivering constantly changing information. Daily
dashboard publication fits this model quite nicely.
Data is normally moved from the datahubs to a user
workstation where it is further processed. Every day after the data has been collected by the
datahubs, a process is launched that parses the files
for a predetermined set of servers, and then creates
• Workstation setup information has been provided
customized CSV files with the basic OS
to users, so they can access and download
measurements (CPU, Load Avg, Memory Util %, Page
desired information to their workstations.
Scans, Network IO). The customized CSV files are
• Provision is made for 1-minute collection where
imported into a database and Perl programs are used

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

to query the data to represent various views (raw,

averaged and trended). Graphs are drawn and saved
to .png files. Weekly charts are created (with daily
updates). HTML is written to present those graphs as

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
static content for the business partners to peruse at
their convenience. Examples are reflected in the
charts in Diagram F.

Using the Data – Performance Analysis and

Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

Troubleshooting

Capacity planning and dashboard graphs are used for

normal performance management, showing peak
periods and highlighting problem areas. Additionally,
the native commands used to capture are the same
ones performance analysts normally use to observe
specific performance parameters. More in depth
analysis is done when the server is sick. When it is
healthy there is no compelling reason to collect more
data.

Results

We have not calculated the cost savings from our

efforts, partly because it is too early to determine the
total benefits as this is still a work in progress.

Our staffing has increased in support of data collection

(four staff vs. two for the third-party product).
However, with the earlier budget/cost restrictions, we
were limited in the number of servers from which we
could collect data. We are now collecting systems
data from over 10,000 servers, and our DNT product is
now a standard that is being installed on all company
midrange servers.

We are now focusing our efforts on further automating

our implementation and monitoring processes;
providing better tools for our capacity analysts, so they
can support a wider client base; and improving our
production and delivery of automatically generated
charts and dashboards.

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Table A - Original Requirements

Meas.

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Metric Groups Interval Metric Unit Description
once per operating system name(sun,hp,aix,
config day Os linux,etc.)
os level operating system level
os version operating system Version
word operating system Word Length
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

length (64/32)
memory KB physical memory
num of
cpu's number of cpu's
kernel parameters (important)
research
cpu speed (important)
hostname hostname
ip address ip address
block device model (nice to have)
nic type, full duplex y/n?, nic speed
(important)
physical swap configed kb
file once/day filesystem name name of filesystem
used kb used kb of filesystem
available kb available kb of filesystem
capacity kb capacity of filesystem
mounted
on mount path
used % % percentage used
network 5 min. name interface name
maximum transmission unit (packet
mtu size)
net/dest attached network
address address for each interface
ipkts pkts input packets per interface
ierrs errs input errors per interface
opkts pkts output packets per interface
oerrs errs output errors per interface
colls collisions collisions per interface
queue bytes send
bytes in / bytes out
% tcp resend
process 5 min. pid name process id
user name user name
cpu % cpu used by process
rss kb resident set size of process in KB
size of process in virtual memory in
vsz kb KB
command name command being executed
cpu 5 min. user % cpu in user mode
sys % cpu in system mode

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

cpu in idle w/process waiting for

wio % block
idle % cpu in idle

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
by individual cpu
average total cpu
load avg. 1,5,15 minutes (or 5,10,15
min - AIX)
portion of time device was busy
block device info. 5 min. busy % servicing a transfer request.
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

average number of requests

avg queue requests outstanding during that time.
number of read/write transfers from
r+w/s reads+writes/sec. or to device
number of bytes transferred in 512-
blks/sec blocks/sec. byte units.
avwait ms average wait time in milliseconds.
average service time in
avserv ms milliseconds.
device name in /dev/format c01
average queue length of processes
queue 5 min. runq-sz number of procs. in memory and runnable.
runq-occ % percent of time run queue occupied
average length of swap queue of
swpq-sz number of procs. runnable processes
percent of time swap queue of
swpocc % runnable processes was occupied

system swapping/
switching 5 min. swpin/s transfers/sec swap ins
bswin/s blocks/sec blocks (512) swapped in

swpot/s transfers/sec swap outs

bswot/s blocks/sec blocks (512) swapped out

pswch/s switches/sec process switches

swap in use in kb
pages per second scanned by the
paging 5 min. pgscan/s pgs/sec. page stealing daemon.
pgs/sec. page ins
pgs/sec. page out

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Table B – Comparison of Commands by Vendor (subset)

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Native Command Grouping Metric Unit Description Sun HP AIX
df filesystem name name of filesystem x x x
used kb used kb of filesystem x x x
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

available kb available kb of filesystem x x x

capacity kb capacity of filesystem x x x
mounted on mount path x x x
used % % percentage used n/a x x
netstat name interface name x x x
mtu maximum transmission unit (packet size) x x x
net/dest attached network x x x
address address for each interface x x x
ipkts pkts input packets per interface x x x
ierrs errs input errors per interface x n/a x
opkts pkts output packets per interface x x x
oerrs errs output errors per interface x n/a x
colls collisions collisions per interface x n/a x
queue bytes send/receive x n/a n/a
sar cpu user % cpu in user mode x x x
sys % cpu in system mode x x x
wio % cpu in idle w/process waiting for block x x x
idle % cpu in idle w/process waiting for block x x x

block device busy % portion of time device was busy servicing a transfer request x x x

avg queue requests average number of requests outstanding during that time. x x x
r+w/s reads+writes/sec. number of read/write transfers from or to device x x x
blks/sec blocks/sec. number of bytes transferred in 512-byte units. x x x
avwait ms average wait time in milliseconds. x x x
avserv ms average service time in milliseconds. x x x
average queue length of processes in memory and
queues runq-sz number of procs. runnable. x x x
runq-occ % percent of time run queue occupied x x x
swpq-sz number of procs. average length of swap queue of runnable processes n/a x x
percent of time swap queue of runnable processes was
swpocc % occupied n/a x x
buffer activity bread/s transfers per second between system buffers and disk or other block devices. x x x
bwrite/s transfers per second between system buffers and disk or other block devices. x x x
lread/s transfers per second access of system buffers x x x
lwrite/s transfers per second access of system buffers x x x
%rcache % cache hit ratios (1-bread/lread) x x x
%wcache % cache hit ratios (1-bread/lread) x x x
pread/s transfers per second transfers using raw (physical) device mechanism. x x x
pwrite/s transfers per second transfers using raw (physical) device mechanism. x x x
-
-
-

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Table B (continued) – Comparison of Commands by Vendor (subset)

page faults per second that are satisfied by reclaiming a
paging activity x n/a n/a

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
atch/s flts/sec. page currently in memory (attaches per second).
pgin/s req/sec. page-in requests per second. x n/a n/a
ppgin/s pgs/sec. pages paged-in per second. x n/a n/a
page faults from protection errors per second (illegal access
pflt/s flts/sec. to page) or "copy-on-writes". x n/a n/a
address translation page faults per second (valid page not in
x n/a n/a
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

vflt/s flts/sec. memory).

faults per second caused by software lock requests
slock/s flts/sec. requiring physical I/O. x n/a n/a
cycle/s cycles reports number of page replacement/sec n/a n/a x
flts/s flts/sec. reports number of page faults/sec n/a n/a x
slots slots reports number of free pages on the paging spaces n/a n/a x
odio/s sec reports number on non-paging diskio/sec n/a n/a x
paging activity pgout/s req/sec. page-out requests per second. x n/a n/a
ppgout/s pgs/sec. pages paged-out per second. x n/a n/a
pages per second placed on the free list by the page
pgfree/s pgs/sec. stealing daemon. x n/a n/a

pgscan/s pgs/sec. pages per second scanned by the page stealing daemon. x n/a n/a
the percentage of UFS inodes taken off the freelist by iget
which had reusable pages asso-
ciated with them.These pages are flushed and cannot be
reclaimed by processes.Thus, this is the percentage of igets
%ufs_ipf % with page flushes. x n/a n/a

Table C – Captured Data on the Agent

/banktools/dnt/collect

-rw-r--r-- 0 20893 14500 0 Jul 2 0:05 cron-er

-rw-r--r-- 0 20893 14500 20538 Jul 2 17:00 <servername>-20050702-df
-rw-r--r-- 0 20893 14500 0 Jul 2 0:00 <servername>-20050702-er
-rw-r--r-- 0 20893 14500 1307040 Jul 2 16:59 <servername>-20050702-io
-rw-r--r-- 0 20893 14500 380275 Jul 2 17:00 <servername>-20050702-nb
-rw-r--r-- 0 20893 14500 1533878 Jul 2 17:00 <servername>-20050702-ps
-rw-r--r-- 0 20893 14500 589000 Jul 2 17:00 <servername>-20050702-sa
-rw-r--r-- 0 20893 14500 48995 Jul 2 17:00 <servername>-20050702-sw
-rw-r--r-- 0 20893 14500 16605 Jul 2 17:00 <servername>-20050702-up
-rw-r--r-- 0 20893 14500 50388 Jul 2 16:59 <servername>-20050702-vm

/banktools/dnt/archive

-rw-r--r-- 0 20893 14500 515714 Jun 26 0:03 <servername>-20050625-linux.tar.gz

-rw-r--r-- 0 20893 14500 516726 Jun 27 0:03 <servername>-20050626-linux.tar.gz
-rw-r--r-- 0 20893 14500 518977 Jun 28 0:03 <servername>-20050627-linux.tar.gz
-rw-r--r-- 0 20893 14500 543040 Jun 29 0:03 <servername>-20050628-linux.tar.gz
-rw-r--r-- 0 20893 14500 534348 Jun 30 0:03 <servername>-20050629-linux.tar.gz
-rw-r--r-- 0 20893 14500 562216 Jul 1 0:03 <servername>-20050630-linux.tar.gz
-rw-r--r-- 0 20893 14500 539197 Jul 2 0:03 <servername>-20050701-linux.tar.gz

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Table D – Native Commands

Native Commands Used

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
outfile AIX Sun HP Linux
cf uname -a uname -a syslog.log cpuinfo
oslevel isainfo -b vmstat -n meminfo
date date
ls -las /unix swap -s
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

no -a netstat -in -f inet

netstat -v psrinfo -v
lsdev -C THEN iostat -En
lsattr -l <devname> -E -H prtconf
prtconf sysdef

df df -kP df -kl df -kl df -klP

io iostat -d iostat -tDc iostat iostat -d

mp mpstat

nb netstat -s -p tcp netstat -s -P tcp -f inet netstat -in netstat -in

ne netstat -in netstat -in -f inet

nk netstat -in -k inet THEN

netstat -k <int> -f inet

pd prtdiag

ps ps -eo pid ps -eo pid ps -eo pid ps -eo pid

user user user user
pcpu stime pcpu pcpu
rss time sz rss
vsz pcpu vsz vsz
args rss args args
time OR vsz
ps auxw args

sa sadc 1 1 sadc 1 1 sadc 1 1 sadc 1 1

sw lsps -a swap -s

up uptime uptime

vm vmstat vmstat vmstat vmstat

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Diagrams E1-E3 – Samples of Charts generated by Capacity Analysts using Excel

E1

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Dot Com Web
250000

200000
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

150000
Spec

100000

50000

0
Apr-04

Jun-04

Aug-04

Oct-04

Dec-04

Feb-05

Apr-05

Jun-05

Aug-05

Oct-05

Dec-05

Feb-06

Apr-06

Jun-06

Aug-06

Oct-06

Dec-06

Feb-07

Apr-07

Jun-07

Aug-07

Oct-07

Dec-07

Feb-08

Apr-08

Jun-08
Month

.COM Utilization (baseline) .COM Utilization (added) ABC Project Total .COM Capacity (spec)
Contingency Capacity (spec) 90% Contingency Capacity 80% Contingency Capacity 70% Contingency Capacity
.COM Actuals .COM Previous Forecast

E2

Dot COM Web Capacity - CPU Utilization

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
Apr-04
May-04
Jun-04
Jul-04
Aug-04
Sep-04
Oct-04
Nov-04
Dec-04
Jan-05
Feb-05
Mar-05
Apr-05
May-05
Jun-05
Jul-05
Aug-05
Sep-05
Oct-05
Nov-05
Dec-05
Jan-06
Feb-06
Mar-06
Apr-06
May-06
Jun-06
Jul-06
Aug-06
Sep-06
Oct-06
Nov-06
Dec-06
Jan-07
Feb-07
Mar-07
Apr-07

.COM Utilization (baseline) .COM Utilization (added) ABC Project

Total .COM Capacity .COM Contingency Capacity 90% Contingency Capacity
80% Contingency Capacity 70% Contingency Capacity .COM Actuals

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

E3

On-line Activity - Concurrent Sessions

180000

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
160000

140000

120000
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

100000

80000

60000

40000

20000

0
Apr-04

May-04

Jun-04

Jul-04

Aug-04

Sep-04

Oct-04

Nov-04

Dec-04

Jan-05

Feb-05

Mar-05

Apr-05

May-05

Jun-05

Jul-05

Aug-05

Sep-05

Oct-05

Nov-05

Dec-05

Jan-06

Feb-06

Mar-06

Apr-06

May-06

Jun-06

Jul-06

Aug-06

Sep-06

Oct-06

Nov-06

Dec-06

Jan-07

Feb-07

Mar-07

Apr-07
User Count Forecast Added Customers Keep Alive Total Model Capacity
Contingency Model Capacity 90% Capacity 80% Capacity 70% Capacity
User Count Actuals Model Session Previous Forecast

Diagrams F1-F7 – Samples of web-based dashboard charts

F1 – Initial Dashboard for Line of Business

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Clicking on ‘OLB Model App’

F2 – Application within LOB

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

Clicking on ‘Click Here for concurrent sessions charts’

F3 – Aggregate for Concurrent Sessions

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Then, clicking on ‘More Detail’ and va2polb030, we get

F4 – CPU vs. Sessions for an individual server

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

If we go back to Diagram F2 and click on the chart for ‘Week of 2005-08-25 to 2005-08-31, and ‘RV’, we get

F5 – CPU Utilization for multiple servers

Find a CMG regional meeting near you at www.cmg.org/regions

 Learn the basics and latest aspects of IT Service Management at CMG's Annual Conference - www.cmg.org/conference

Once again, clicking on va2polb030, then on ‘LOADAVG’, we get

F6 – Load Average for a single server

Buy the Latest Conference Proceedings and Find Latest Computer Performance Management 'How To' for All Platforms at www.cmg.org
Join over 14,000 peers - subscribe to free CMG publication, MeasureIT(tm), at www.cmg.org/subscribe

Finally, clicking on ‘DISKUTIL‘ yields the following chart

F7 – Disk Utilization for a single server

Find a CMG regional meeting near you at www.cmg.org/regions