|
Archives
Connecticut Computer
Measurement Group
Autumn CCMG Membership Meeting Agenda
Radisson Hotel, Cromwell - Friday, October 18, 2002
Every year in October, CCMG has the opportunity to showcase at least
one, (usually more) session that will be presented at the National
CMG conference in December. This year we have three such sessions
with a wide range of appeal; this includes a medley of sysplex performance
tuning, more efficient statistical monitoring, and getting into the
web laboratory. CCMG is also proud to host a “home grown”
speaker who will discuss how he developed a simple methodology for
identifying when applications will exceed the batch window. These
are all complemented with a case study illustrating an approach to
managing risk and change in our network world and a discussion on
WebSphere performance monitoring. Connecticut CMG is happy to present
these sessions to our membership.
Agenda:
08:00-08:30 Registration and Coffee
08:30-08:45 Introductory Remarks
Open Systems Track
08:45-10:00 HOW TO: Prepare Effectively for Network Outages and Traffic
Growth
Best Practice Approaches for Managing Risk & Change
Todd Kaloudis - OPNET
Managing networks that meet the needs of changing business conditions
is difficult. In this session we will present case studies that illustrate
how this can be done without significantly increasing infrastructure
Total Cost of Ownership. Visualize the current network, including
traffic/utilization Capture detailed information to understand the
impact of specific application deployments Model the effect of traffic
growth on routing and end-to-end performance Analyze “what if”
scenarios to develop failure resilient networks with acceptable service
levels
------------
10:00 Break - Refreshments provided by OPNET
10:15 – 11:30 Achieving Maximum Uptime with WebSphere Performance
Monitoring
Barry Lamkin - Candle
If your company is implementing WebSphere, then there is a pressing
need to monitor the WebSphere Application Server. This session will
talk to the risks and exposures from outages and slowdowns. It will
discuss different approaches for management and monitoring, and present
an understanding of the entire WebSphere environment.
------------
11:30- 12:30 Lunch (Provided) – Hot Buffet
Mainframe Track
8:45-10:00 Forecasting Batch Window Requirements
Dennis Rembiesa - CIGNA
This session will take you through the data collection, download,
and model development necessary for estimating batch window growth.
This process was developed holistically during a requested study of
a troublesome application, and developed into a sophisticated automated
model with an easy-to-use GUI front-end built with Microsoft Access.
Dennis will use real examples, demonstrate how the model works, and
discuss line-of-business acceptance of the results.
------------
10:00 Break - Refreshments provided by OPNET
10:15-11:30 Sysplex Performance Tuning Medley
Gary King - IBM
A variety of topics on the performance of IBM’s Parallel Sysplex
will be presented. Topics to include: performance sensitivities overview,
coupling facility options, link technology options, how to determine
what techology is right for your sysplex, and the value and costs
associated with current and future coupling facility duplexing.
------------
11:30- 12:30 Lunch (Provided) – Hot Buffet
Combined Sessions
12:45-2:00 Statistical Performance Monitoring
Yiping Ding - BMC
Traditional performance monitoring methods have an overhead that
is proportional in both space and time to the spill rate. They are
not scalable in a distributed computing environment of hundreds or
even thousands of managed nodes; the large quantity of data collected
cannot be pushed or pulled efficiently across the network for displaying/reporting.
In this session we propose a few scalable alternatives such as only
sending and displaying (at a lower frequency) essential data/information
(mean standard deviation exceptional values etc.) for each metric.
------------
2:00 Break - Refreshments provided by OPNET
2:15-3:30 The Web Laboratory - Performance on the Web
Denise Arruda - GTECH Corporation
As end users are doing more business on the web and expecting faster
response, more businesses are looking to understand and test the performance
of their web applications. In order to gain this understanding, more
businesses are establishing application performance teams. Performance
teams work with development and management teams to define performance
requirements, and understand expected transaction load, so that the
application can be tested in a laboratory environment. This paper
will describe the lessons learned while establishing an application
performance team and laboratory.
------------
Return to Top
|
