|
Session Abstracts
Keynote
sessions are designed to provide relevant topics that apply to all
attendees of the Application Performance Summit, both business and
technical.
Keynote
1:
Predicting the Impact of Change
Noel
Barnard,
President
and CEO, HyPerformix
In Mark Lutchen’s book, Managing IT as a
Business, the key theme throughout is that if you’re going to
manage IT as a business, you must first be able to manage change …
HyPerformix takes “managing change” to the next level with its
predictive performance and capacity management solutions and
conviction that if you’re going to effectively manage change, you
must first be able to predict the outcome of change.
Ms.
Barnard talks to the value of being able to predict the impact of
infrastructure, application, and workload change and how being
proactive vs. reactive reduces risks, eliminates unnecessary spending,
shortens time-to-market, and ensures performance and availability of
mission-critical applications.
Speaker
Biography
Ms. Barnard joined HyPerformix in 2000 and brings to
the company over twenty years of business and management experience in
the Information Technology industry. A results-oriented leader
with strong financial and corporate expansion experience, Ms. Barnard
offers HyPerformix a well-balanced portfolio of skills, including
general and financial management, operations, sales/marketing,
business development, strategic planning, and professional services.
Prior
to HyPerformix, Ms. Barnard was Vice President of Professional
Services, Tivoli Systems (an IBM company); National Practice Leader,
IBM Global Services (where she served on the Enterprise Systems
Management Core Team); Senior Vice President of Business Development,
Snyder Communications; Senior Vice President of Sales and Operations,
AMR Services, an AMR (parent to American Airlines) subsidiary; and
Vice President of Sales and
Marketing, The Sabre Group (also an AMR subsidiary).
Keynote
2:
Align,
Automate and Optimize Your Business with Predictive Capacity
Management
Jan-Maarten
van Dongen,
CTO
OpenView, HP
During the past few years, the number of HP
OpenView customers wanting to discuss best practices related to
‘Capacity Management’ has continued to increase. However,
every company seems to have a different interpretation or approach for
managing capacity, including:
-
measuring server
performance
-
building data
warehouses
-
performing
simple trending & forecasting
-
building
predictive models
-
testing
transaction workloads
-
developing ITIL
Capacity Management practices
-
and many more…
For the most part, companies are interested in
capacity management as related to current or future activities such as
optimization, cost reduction, resource utilization, process
automation, infrastructure consolidation and virtualization.
Companies
would like to know more about how all these issues relate to one
another and – more importantly – how to achieve the greatest
business value through Capacity Management. Learn what
Hewlett-Packard, including HP Labs and HP OpenView are doing to help
companies introduce best practices in all facets of Capacity Manager,
both today and into the future.
Speaker
Biography
Mr. van Dongen is an internationally recognized leader
and expert in Management Software. As Chief Technology Officer for the
OpenView Software Business Unit (OVBU), Jan-Maarten is responsible for
the Hewlett-Packard
“Management for the Adaptive Enterprise” strategy, architecture
and future directions. He drives the OpenView solution and portfolio
development, partner and strategic alliance programs.
Jan-Maarten
van Dongen joined HP in 1985. Prior
to his CTO role he was the worldwide OpenView chief architect, leading
the overall HP enterprise management solution’s architecture and
integration. Before that he was the specialist in Benchmarking,
Performance Analysis/Modeling and Architecting Mpe, UNIX, Windows and
Network based computer solutions in HP’s consulting organization.
Jan-Maarten van Dongen is Distinguished Technologist
within the Hewlett Packard Company and holds an Engineering degree in
Electronics from the
Technical
University
in Delft, The Netherlands.
Keynote
3:
Leveraging
Performance Engineering Assets - Strategic
Enterprise
Optimizer Models
Dr.
Charles Letner, Senior
Manager of Performance and Capacity Modeling, Alltel
Performance engineering (PE) using modeling is an
accepted approach for resolving performance issues.
The typical approach is to execute a PE project to answer
specific questions with model creation a key activity.
The models are traditionally created from scratch for each
project and viewed as a tactical deliverable useable primarily in the
context of the project. An
alternative approach is to view models as strategic assets.
In this approach models have a life that extends far beyond any
specific project. When
shifting to this paradigm the concept of an enterprise model becomes
possible. In the context
of businesses that are run on a highly integrated computing
environment the enterprise model concept is extremely powerful.
When this concept is combined with modeling from production a
significant increase in capabilities can be realized.
Further, the significant knowledge inherent in the models can
be institutionalized.
This
presentation will describe the concept of an enterprise model and
present a concrete example created in Optimizer.
Examples of how the enterprise model has been used to answer
questions will be discussed. The
process and techniques required will be described allowing you to go
back to your company and begin to implement the enterprise modeling
approach. Finally, the
value-add proposition of enterprise Optimizer models will be discussed
to provide motivation for adopting the approach.
Speaker
Biography
Dr. Letner is a senior manager of performance and
capacity modeling at Alltel Communications, Inc.
Prior to the capacity and performance work he was a senior
software and hardware infrastructure architect focusing on
scalability, availability, and reliability. He
entered the IT industry as a software developer and spent four years
writing and designing applications in C++ and Java.
In 1996 he completed a Ph.D. in biochemistry specializing in
computational biochemistry. As
a result of this diverse background he brings a strong sense of
business and technical semantics to modeling.
This combined with rigorous data analysis and statistics result
in a high degree of success in model validation and the forecasts
derived from those capacity and performance models.
Alltel
is a customer-focused communications company with more than 15 million
customers in 36 states and nearly $10 billion in annual revenues.
Keynote
4:
Proud To Be Different - Performance Engineering
at Nationwide Building Society
Andrew
Thompson,
Senior
Systems Consultant, Nationwide Building Society
Nationwide
Building
Society is the largest building society in the world. As a mutual
organisation, it does not suffer from divided loyalties between
customers and shareholders. This is what makes it different from other
UK
financial organisations. To provide the most benefit to its members,
Nationwide needs to react quickly to changing market opportunities and
financial regulations. The challenge for the technology division is to
provide optimum business services that don't just function, but also
provide superior performance; now and in the future.
Find
out how Nationwide has used HyPerformix methodologies and tools to
assist in the delivery of tactical models for key business services.
Also, learn about Nationwide's goal to build on this success, by
delivering an effective Performance Engineering service to predict the
impact of change, and deliver best value to our members.
Speaker
Biography
Mr. Thompson
joined the Nationwide Building Society, Capacity Management team
straight from University.
In
the 18 years since joining the society as a trainee, Andrew has risen
to be the society’s Capacity Management Consultant responsible for
the development of monitoring & modelling technologies, processes
and capacity management strategies. In that time Andrew has worked at
capacity planning across all aspects of the society’s technical
infrastructure and was instrumental in developing the HyPerformix
relationship originally based upon the society’s use of Workbench
and now IPS. Andrew holds a degree in Physics and Electronics.
Keynote
5:
Building a Capacity Management Practice
Beth
Hage,
VP
Retail Strategic Performance Testing, JP Morgan Chase
More and more organizations are responding to
pressure to decrease technology costs without affecting the customer.
In the good old days, every application team had its own
hardware, safe from the rogue processes of other applications.
With average hardware utilization topping off at 20%, and with
skyrocketing costs to host and maintain systems, the golden age of
application silos is over.
The
challenge we are faced with is reducing the dollars spent on hardware
up-front and maintenance costs without fear.
In order to do this, a disciplined approach must be developed
to predict system capacity and performance and to alert the
appropriate parties in advance of capacity problems. This presentation
will discuss the challenges and rewards of building out your own
Capacity Management practice.
Speaker
Biography
Elizabeth Hage is Vice President of Retail Strategic
Performance Testing at JP Morgan Chase, a leader in investment
banking, financial services for consumers and businesses, financial
transaction processing, asset and wealth management and private
equity.
Having joined JP Morgan Chase in October 2002, Ms. Hage
has spent more than 10 years building, maintaining, and optimizing
large distributed systems.
Elizabeth
’s team provides technology consulting services for JP Morgan Chases
Retail Line of Business. This
group is an elite provider of end-to-end performance testing, capacity
planning, and monitoring services to over 500 applications inside and
outside the Retail line of business.
Ms.
Hage has a B.S. in Communications from the University
of
Illinois, Champaign-Urbana and is a Certified Software Quality Analyst
(CSQA).
Keynote
6:
The
Evolution from System to Business Service Management
Bob
Yellin,
President,
Z-Bot
To
improve their responsiveness to business needs, IT organizations are
changing the way they manage their infrastructure and application.
Traditional systems management has been focused on the performance and
availability of silo components. Today’s leading organizations have
made 2 fundamental shifts: 1) the introduction of SLM solutions has
enabled IT to manage from an end-user perspective and 2) the
introduction of Business Service Management solutions has allowed IT
organizations to optimize the alignment of IT assets to business
needs. This session explores the people, process and technology
changes IT organizations must address as they mature to adopt BSM
solutions. The session will also look at the role of predictive
modeling to support the optimization of both quality of service and
cost of service delivery.
Speaker
Biography
Mr.
Yellin has been widely recognized as a thought leader in the systems
management market. Bob has more than twenty years experience
spearheading development, strategic planning and technical management
initiatives with Tivoli, BST Software's ENDEVOR
and Legent Corporation's overall product and technical strategies.
Keynote
7:
Managing the Business
of IT
Michael
McCarthy, Director
of Strategy, Tivoli
The ultimate goal of capacity planning or any
other systems management activity is to serve the business. IT
operations today are under great pressure not only to improve service
and to reduce costs but to be seen to be doing so. The best response
to this challenge is to implement management processes and automate
these processes as much as possible. In this session, we will
introduce IBM's IT Service Management solution and show how customers
can use this to build process management automation to maximize
operational effectiveness. We will also cover the need for capacity
planning as a key step in a number of critical processes and show how
this integration may be achieved.
Speaker
Biography
Mr.
McCarthy is the Director of Market Management for IT Service
Management, IBM Tivoli Software, where he has worldwide responsibility
for the business strategy for IBM's IT Service Management products.
Mike joined IBM in 1986 as a software engineer for IBM network
management software. He has held a variety of positions with
IBM, including: manager of channels for IBM Networking Software in
Asia Pacific, market strategist for IBM Retail Store Solutions, and
Director of Market Management for IBM TotalStorage Open Software.
He also worked for Medic Computer Systems, a healthcare software
company, where he was responsible for marketing strategy and analyst
relations. Mike has a bachelor’s degree in computer
engineering from
Michigan
State
University
and a master’s degree in business administration from Duke
University.
Keynote
8:
Rapid Fire Capacity
Management
Torea
Rodriguez,
Manager of Performance and Capacity Management, eBay
Over the span of 10 years, eBay, Inc. has been in
rapid growth with no plateau in sight. Torea Rodriguez speaks to
Capacity Management in the face of such rapid change and growth. Key
focuses in this discussion will be on the people, process, tools and
organizational alignment.
Speaker
Biography
Ms.
Rodriguez has 10 years of management experience in the Silicon Valley
Internet business space. Over the last 5 years she has taken a keen
interest in performance engineering and capacity planning. Since
coming to eBay, Inc. in 2002, she has focused on the Capacity
Management of the Marketplace platform. Founded in 1995, eBay enables
ecommerce on a local, national and international basis with an array
of websites -- including the eBay Marketplace, PayPal, Skype, Kijiji,
Rent.com and Shopping.com -- that bring together millions of buyers
and sellers every day.
Keynote
9:
Performance and Capacity
Optimization for SAP
Barry
Weber,
Managing Director, Genilogix
Many organizations are looking for clarity on the
value and relationship of performance testing and modeling.
In this presentation, some “successful” SAP®
performance testing case studies will be reviewed.
They will be analyzed for achievement of goals, value and best
practices. A new
HyPerformix offering, Optimizer for SAP, will be presented.
Best practices for coupling performance testing and modeling
will be provided that would be interesting to all customers including
those that are considering upgrades or changes to SAP.
Speaker
Biography
Mr. Weber is a managing director with Genilogix LLC, a
professional services firm that focuses on quality assurance,
performance engineering and IT governance.
He leads the performance engineering practice.
Mr. Weber brings more than 25 years experience in IT
management, IT process improvement, information technology
architecture, program management, EAI development, big-4 professional
services delivery and high volume ecommerce web site infrastructure
engineering and management.
In his career, he has lead development, operations and
technical infrastructure for Barnes & Noble.com, led the US
strategy services for Mercury Interactive, managed a practice area for
PwC on EAI integration to SAP, and consulted to Fortune 100 IT
organizations on implementing centers of excellence for quality,
performance engineering, and operations.
Keynote
10:
Achieving Better Application Performance
Terry
White, EDS
Fellow and Enterprise Architect, EDS
Terry
White will discuss EDS' experiences using HyPerformix,
and modeling & simulation techniques, to understand system
performance and behavior, looking for potential problem areas
that might be overlooked using conventional methods. Mr.
White will use two case studies from past projects to
illustrate the approach and results as well as touch on other related activities
at EDS.
Speaker
Biography
Mr.
White, an EDS Fellow, is an IT Architect with over 20 years experience
in the IT and manufacturing industries. The title EDS Fellow is
awarded to the corporation’s most innovative thought leaders in
recognition of their exceptional achievements. As an EDS Fellow, White
helps develop enterprise-wide initiatives that shape the future of
EDS. His current responsibilities include assisting clients in
developing business and technology solutions that provide strategic
value; assisting EDS and clients in developing their technical
strategy and technical direction; working on EDS, partner, and client
teams to ensure operational excellence of implemented solutions;
mentoring team members; and providing oversight and thought leadership
to EDS and its clients.
Keynote
11:
Using Modeling to Optimize
Disaster Recovery Planning
Brian Zawada, Director of Consulting Services,
Avalution Consulting
Many
organizations have made investments in disaster recovery plans and
dedicated disaster recovery infrastructures, but there are still
significant risks in current disaster recovery approaches. Sizing of
the disaster recovery infrastructure is often “rule of thumb”,
with little understanding of what infrastructure is actually needed.
Testing validates that data can be restored, and that the
infrastructure can execute a set of transactions, but performance
testing of disaster recovery infrastructure under actual or
anticipated loads is rarely done. This session looks at the current
strategies companies are using to manage their disaster recovery
infrastructure. The session will also explore how modeling can be used
to validate the performance of existing disaster recovery plans, as
well as provide a method to optimize disaster recovery investments.
Speaker
Biography
Mr. Zawada is the Director of Consulting Services
for Avalution Consulting, a firm specializing in event risk management
and business continuity solution design, development, implementation
and long-term maintenance. Brian has been involved in the business
continuity industry for ten years and is a member of the Continuity
Insights Editorial Advisory Board and President of the Northern Ohio
Association of Contingency Planners (ACP) chapter.
Track A: HyPerformix
Product Overview
A1:
Getting
the Most out of Data Manager - Collecting, Normalizing and Analyzing
Performance Data
Craig
Kim, CTO, Octo
Performance Engineering, Inc.
Data
Manager is the data collection, normalization and warehousing
component to IPS Capacity Manager. Beyond the pretty face of Data
Manager, rich sets of business data are available to be mined and
utilized to maximize the ROI. Mr. Kim discusses where they can be
found, what can be mined and analyzed, and how they can be accessed
from various tools and facilities.
Speaker Biography
Mr.
Kim is CTO of Octo Performance Engineering, Inc., which specializes in
providing IT performance engineering services to telecommunications
and banking industries. While he was with AT&T Wireless, he
founded and championed the performance engineering practice to resolve
the existing performance issues and to predict the scalability of CRM
and billing systems on Oracle. He received a Bachelor of Chemical
Engineering from the
University
of
California,
Berkeley, and an MBA from the
University
of
Washington.
A2:
Introduction
to IPS Performance Optimizer
Rob
Carruthers,
Director of Product Management, HyPerformix
HyPerformix
recently released a series of major enhancements to Performance
Optimizer. This session will focus on giving you what you need to know
about these new features. They include scenario cost optimization, a
new Navigator for HP OpenView, modeling support for shared storage
devices, and enhancements to common component models.
Speaker
Biography
Mr.
Carruthers serves as Director of Product Management for HyPerformix
Inc. He
has over 20 years experience in the software and computer systems
industry, having worked for Hewlett Packard, Convex Computer
Corporation, and Harris Corporation. During
his tenure at HP he had responsibility for the development and market
introduction of HP's highest performance server product line, the HP
9000/X class. Mr. Carruthers holds a degree in Mechanical Engineering
from
Texas
A&M
University
and an MBA from the
University
of
Texas
at Dallas.
A3:
Workload
Gathering and Analysis
Dawn
Murphy, Proactive Performance Engineer, State Farm
In
order to evaluate the impact of work on a system one must understand
what that work is. This workload analysis phase is often rushed and
the results oversimplified to a simple transaction rate. Often our
test hardware capacity is smaller in test than in production and each
tier may have a different size ratio of test hardware to production
hardware. This presentation will demonstrate a systematic workload
gathering and analysis procedure. We will capture a ‘picture’ of
the work done in a system and modify that work to fit our test system.
Situations that require modeling are identified and mitigated. Our end
result is an understanding of the user demands on the system and a
documented workload characteristic.
Speaker
Biography
Ms.
Murphy
has worked for State Farm for 5 years in the Systems Department,
spending the past 2 years in the Performance area as a Proactive
Performance Engineer. This
entails creating analytical and simulation modeling for projects as
well as pre-project consulting. Dawn
attended and passed the Enterprise Performance Engineering
Certification (EPEC) course given by HyPerformix and attended both the
HyPerformix IPS and the Advanced Data Collection and Analysis
courses. Previously at
State Farm she has implemented web based knowledge management
applications in various call centers.
Before joining State Farm,
Dawn Murphy
worked at Larson’s Texts in
Erie
,
Pennsylvania
in web development and programming.
This included creating dynamic educational web based content as
well as back office programming and web server administration.
Dawn earned her BS in Computer Science at
Edinboro
University
in Pennsylvania. In her free time she
enjoys running, biking, and volleyball.
A4:
The
Online Bank Model - Lives!
James Thring, Senior
Consultant, Nationwide
Building Society
In
2002 Nationwide Building Society created its first performance
engineering model for their internet banking service. Numerous
expansions to this service resulted in significant cost and effort to
re-build the model in 2004. This lesson learned, it was decided that
it would be far more efficient to maintain, rather than re-build, a
performance model. Find out about the challenges and benefits of
keeping models alive and kicking.
Speaker
Biography
Mr. Thring
has 10 years experience in the IT industry. He spent 6 of these years
working for McKesson (UK), as a network professional, providing IT
solutions to the health market.
Moving to Nationwide Building Society, James joined the
Capacity Management team, where he has been able to gain better
understanding of the impact of infrastructure configuration on
application performance. Using the HyPerformix IPS suite, James has
built several models, and continues to develop a formal Performance
Engineering service in the organisation. James holds a degree in
Science, and a post-graduate diploma in Computer Science.
A5:
Integrating
Tivoli
ITM and IPS Capacity Manager for Predictive Performance Management
Peter Marshall, IBM Tivoli
Market Manager, Tivoli and Derek Weeks, VP Product Management,
HyPerformix
IPS
Capacity Manager is certified "Ready for
Tivoli
" and IBM Tivoli recently signed on as a global reseller of this
IPS product. For years, IBM
Tivoli
Monitoring (ITM) has alerted users to encroaching or critical
performance issues. ITM has also provided the ability to analyze
historic data to improve visibility to performance trends in the
end-to-end infrastructure.
Tivoli
and HyPerformix will show you how to leverage this historical
performance data for predictive capacity planning and prescriptive
change management. This session provides an overview of how to use
Tivoli performance data in conjunction with HyPerformix's IPS Capacity
Manager to:
-
accurately
model server consolidation efforts;
-
accurately
model workload reassignment;
-
proactively
determine just-in-time procurement of new infrastructure;
-
safely
and quickly model performance and capacity what-if scenarios
without disturbing the production environment; and
-
model
the cascade effect of single/multiple infrastructure changes on
performance and capacity over time.
This
session provides an overview of IPS Capacity Manager, how it
integrates into ITM and how it can be used in common scenarios to
ensure faster problem resolution and fewer problems in the first
place!
Speaker
Biography
Mr.
Marshall is responsible for market management and business strategy
for
Tivoli
's service delivery management product area. He is focused on a number
of emerging business areas and has been working with HyPerformix on
enabling customers to integrate Tivoli
management metrics into capacity management tools. Prior to joining
Tivoli, Marshall
worked in a number of development and strategy roles at Candle
Corporation.
A6:
Optimizing
Performance throughout the Application Life Cycle
Paul
Porter,
Senior
Sales Engineer, Mercury
Poor
application performance and system downtime in production have a
serious impact on an enterprise's top and bottom line. Unfortunately
most IT teams are not organized properly to meet this challenge due to
siloed teams across development and operations, time crunch
and the volume of applications. In order to meet the performance
and availability requirements of the business, IT needs to prevent
performance problems from happening and if they do happen, know about
them as quickly as possible and determine the root cause. This session
explores how integrated
solutions from Mercury and HyPerformix help provide a complete
solution to the performance and availability needs of enterprises.
Speaker
Biography
Mr.
Porter
is
a Senior Systems Engineer for Mercury.
Paul is an expert on Mercury’s performance management
solutions, spanning from Performance Testing, Diagnostics, Capacity
Planning and into Production.
A7:
Modeling
Virtualization of Servers with nWorks and VMWare
Hank
Wright, Executive Vice President, nWorks
VMware
ESX Server has lead the x86 virtual infrastructure evolution from
discreet x86 systems where hardware utilization ranged from 11% –
18% to the virtualized infrastructure where hardware utilization has
been raised to 70% to 80% resulting in huge financial savings and
increased availability. This session will discuss the VMware ESX
Monitor architecture with a focus on performance measurement and event
management. The integration of ESX Server performance, event, state,
and configuration information into enterprise management platforms
such as HP OpenView Operations and Microsoft Operations Manager is a
key aspect of ITIL-based service management designed to maximize
reliability, availability, supportability, and manageability of ESX
Servers.
Speaker
Biography
Mr.
Wright an Executive Vice President at nWorks. A 25-year veteran of the
high tech industry, Wright is responsible for developing, marketing
and supporting nWorks products that include the nWorks Smart Plug-in
for VMware on HP OpenView and the VMware Events Management Pack for
MOM 2005. VMware ESX Server performance monitoring is a key aspect of
these solutions which can provide raw performance data for capacity
planning with the HyPerformix IPS product family. Wright has deep
experience on both sides of the IT market. As the Director of Network
Service at DirecTV he had operational responsibility for all aspects
of the networking and x86 computing infrastructure. Prior to that,
Wright worked for HP as a Senior Solution Architect working with
Hewlett-Packard’s strategic accounts to address their networking
operational requirements.
Track B: HyPerformix
IPS Performance Optimizer Modeling
B1:
Stepwise
Refinement - from Capacity to Performance Models in Optimizer
Richard Gimarc, Director of Technical PE Services, HyPerformix
This
session will describe a new methodology for quickly developing a
capacity planning model during the Test Phase. We will take the tried
and true T-Test Methodology, stand it on its head, and show you how
to get faster results by starting with T3. The intent of
the new T3 methodology is to quickly create an initial capacity
planning model using only T3 test results. You then refine
that model by incrementally including results from the standard T1-T2
tests for select business functions. The result is that you get
a usable model quicker, and you then have the freedom to drill
down with additional refinement using the standard T-Test Methodology.
Speaker
Biography
Mr. Gimarc has over twenty-five years of technology
experience in computer performance analysis and software development.
Mr. Gimarc has broad experience in solving a variety of complex
technology and performance problems affecting a wide range of
applications and computing platforms ranging from standard business
applications, to real-time systems, to operating systems, and
Web-based applications. Mr.
Gimarc has presented numerous times at technical conferences and has
authored several technology papers related to performance engineering.
B2:
Optimizer
Tips/Techniques - The One Hour Model
Richard
Gimarc, Director of Technical PE Services, HyPerformix
and Amy Spellmann, VP and Practice Director, HyPerformix
Have
you ever built a model in an hour? Or does it always take 4-6 weeks?
Join us for a workshop where we build a model in Optimizer in one
hour. We'll start with a 1-page description of the application and its
desired performance characteristics. Using AMG and Modeler, we'll
construct the model and use it to evaluate scalability and performance
of the application. This session will go a long way to dispel the
prevailing attitude that "it takes too long".
Speaker Biographies
Mr. Gimarc has over twenty-five years of
technology experience in computer performance analysis and software
development. Mr. Gimarc
has broad experience in solving a variety of complex technology and
performance problems affecting a wide range of applications and
computing platforms ranging from standard business applications, to
real-time systems, to operating systems, and Web-based applications.
Mr. Gimarc has presented numerous times at technical
conferences and has authored several technology papers related to
performance engineering.
Ms.
Spellmann, VP & Practice Director Performance Engineering Services
for HyPerformix, managing the PE Services business in the Western and
Southern United States
. Amy has 17+ years at HyPerformix, and over 20 years total industry
experience in Software Engineering and Performance Analysis. Amy was
the initial founder of the IT Performance Engineering consulting
practice at HyPerformix, and has modeled hundreds of applications for
HyPerformix clients in many industries, including Financial, Banking,
Insurance, Defense, Hardware Manufacturers and Retail. Amy is a
recognized and popular speaker at association groups such as Computer
Measurement Group, and has published over 10 articles and white papers
with IEEE and CMG.
Amy has a B.S. from Texas
State
University.
B3:
Navigator Template Development & Customization
Allan
Clarke, Senior Engineer, HyPerformix and Tim Wise, Senior Engineer,
HyPerformix
IPS
Performance Optimizer includes Navigator, which is used to automate
and manage methodologies in performance and capacity management.
HyPerformix markets a variety of pre-built methodology templates
targeting specific phases of the application life cycle and specific
data sources. This session explores how Navigator can be used to
customize these templates, or create new templates that are specific
to your own methodologies and business needs.
Speaker
Biographies
Mr. Clarke
has been developing large desktop applications for over 20 years. He
has been with HyPerformix since its inception, and is currently the
lead developer for the Optimizer product suite. His history with the
company includes development on the general purpose simulation tool
SES/workbench, the SES/Objectbench CASE tool, and Strategizer. Mr.
Clarke obtained a BSEE and MSEE from the University
of
Texas.
Mr. Wise
is an 18-year veteran of HyPerformix and has worked both as software
developer and as a performance services consultant. His experience
includes computer system performance analysis, simulation and modeling
of complex systems, and database systems. Tim received a
Bachelors of Science in Computer Engineering and a Master’s of
Science in Computer Engineering from the University
of
Florida
and did doctoral work in Computer Science at the University
of Texas.
B4:
Advanced
Data Analysis – Making sense from nonsense
Keith
Smith, Director of Product Education, HyPerformix
As
anyone who has developed a model knows, data collection and management
are the most important aspects to ensuring the quality of the model.
This session explores the various types of data that is required in a
performance engineering project through various stages of the
application life cycle. It includes tips and techniques to verify
correctness of data and to prepare the data for use in constructing
and validating a performance model of your application.
Speaker
Biography
Mr. Smith
is the Director of Product Education at HyPerformix. Keith leads the
development of all training materials and courses at HyPerformix as
well as his role in classroom instruction.
Keith has over 20 years experience in modeling,
analysis, and engineering in a wide variety of environments. He has
extensive experience in aerospace and defense environments as well as
in high technology industries, leading modeling and analysis and
systems engineering organizations. Keith has spent eight years as a
consultant, leading a mix of technical design, systems integration,
and process reengineering engagements. For the last three years, Keith
has led HyPerformix certification and training efforts.
Keith
has a bachelor’s degree in Applied Mathematics from Carnegie
Mellon
University, and a master’s degree in Operations Research from the
University
of
Texas
at Austin.
B5:
Predicting
Production Performance with IPS Performance Optimizer: The Importance
of Model Variability
Dr.
Charles Letner, Senior
Manager of Performance and Capacity Modeling, Alltel
An important aspect of managing performance is
managing response time extremes, particularly transactions with longer
response times. To
effectively manage the extremes requires an understanding of response
time distributions. IPS
Optimizer discrete-event simulations are well suited to providing
insight into response time distributions.
In this presentation a case study of a CORBA services tier will
be used to demonstrate techniques to add variability to simulations.
The goal is to obtain simulation response time distributions
that match production distributions.
As validation, the results from simulations will be compared to
production response time data. The
results obtained will be discussed in the context of typical
performance problems to demonstrate the importance of variability in
IPS Optimizer models used for performance management.
Speaker
Biography
Dr. Chuck Letner is a senior manager of performance and
capacity modeling at Alltel Communications, Inc.
Prior to the capacity and performance work he was a senior
software and hardware infrastructure architect focusing on
scalability, availability, and reliability. He
entered the IT industry as a software developer and spent four years
writing and designing applications in C++ and Java.
In 1996 he completed a Ph.D. in biochemistry specializing in
computational biochemistry. As
a result of this diverse background he brings a strong sense of
business and technical semantics to modeling.
This combined with rigorous data analysis and statistics result
in a high degree of success in model validation and the forecasts
derived from those capacity and performance models.
Alltel
is a customer-focused communications company with more than 15 million
customers in 36 states and nearly $10 billion in annual revenues.
B6:
Modeling
Virtualization
David Dupre, Performance
Engineer, HyPerformix and Richard
Gimarc, Director of Technical PE Services, HyPerformix
Virtualization
can be used to significantly reduce the number of physical servers and
improve the utilization of existing servers. However, the logical
consolidation of multiple servers onto a single physical server adds
an additional layer of complexity to performance and capacity
management. This session explores how IPS can be used to model
performance and capacity on virtual servers, allowing you to maximize
asset utilization while managing risk.
Speaker Biographies
Mr. Dupre is a Performance Engineer with HyPerformix
who has been in multiple virtualization modeling projects.
Mr. Gimarc has over twenty-five years of technology
experience in computer performance analysis and software development.
Mr. Gimarc has broad experience in solving a variety of complex
technology and performance problems affecting a wide range of
applications and computing platforms ranging from standard business
applications, to real-time systems, to operating systems, and
Web-based applications. Mr.
Gimarc has presented numerous times at technical conferences and has
authored several technology papers related to performance engineering.
B7:
Mine
the Hidden Value from your IT Models
Ken
Zink, Chief Architect, HyPerformix
Significant
resources may be expended in the development of performance models of
IT systems – usually to address a specific tactical issue.
There may be intellectual gold in those models waiting to be
exploited; e.g., application flow, application footprint, and resource
requirements of primary business drivers.
This information is often costly to acquire and should be made
available outside of the modeling and capacity planning organizations.
Models also often have potential strategic value for reuse in
models of similar or related systems or application re-design and cost
trade-off analysis.
The strategic value can be exposed through reusable model
components, sensitivity analysis and cost analysis of alternative
infrastructures.
Speaker
Biography
Mr.
Zink has over twenty years of experience in technology development as
well as the delivery of consulting and professional services
throughout the
United States.
Mr. Zink joined HyPerformix in 1992 as part of the technology
development organization.
Mr. Zink’s experience includes research into massively
parallel computer systems, predictive simulation modeling of systems,
and architecting hardware and software solutions to customer problems.
Mr. Zink was a key contributor to the formation and growth of
the HyPerformix professional services organization and is currently
responsible for the architecture of HyPerformix technology.
Mr. Zink received a Bachelor of Arts degree in Math and Physics
from Kansas
Wesleyan
University.
Track C: The
Strategic Value of Predictive Performance and Capacity Management
C1:
Management -
Getting
the Support Infrastructure Right the First Time
Hadyn
Innis,
ITSM Practice Director, Inventes, Inc
Over eighty percent
of organizations adopting ITIL begin with the Service Support
Processes. The Service Delivery Processes, the proactive processes of
ITIL and the strategic view of any infrastructure present a set of
processes which when properly implemented and made into repeatable and
consistent practices provide a multitude of quantifiable benefits.
This
discussion focuses on Capacity Management- one of the key Service
Delivery Processes and how when applied in a methodical manner
provides a sound basis for current and future infrastructure demand
planning and provisioning execution.
Speaker
Biography
As the Director of
Inventes Institute, Hadyn is responsible for Inventes educational
services and for leveraging thought-leadership initiatives for
strategic selling. As an ITSM Practice Director, Hadyn is responsible
for the leadership and management of ITIL Service Managers in the
Inventes Corporation.
Prior to joining
Inventes, Hadyn was with IBM for more than 15 years. Hadyn has over
eighteen years of experience in information technology, and currently
holds nineteen worldwide patents in the areas of software development,
multimedia, networking, and, most recently Internet based
technologies.
As
an active member of the IT Services Management Forum International
Hadyn is playing a central role in defining the new frameworks for
ITIL3. He is a certified ITIL examiner and on the Certification and
Membership committee of ITSMF USA
and also the past President and Founder of ITSMF USA Central Texas
Local Interest Group.
C2:
Panel
Discussion - Best Practices in Capacity Management throughout the
Application Life Cycle
Jim
Whalen
,
Component Manager, State Farm Insurance
How
are today’s leading companies integrating the Capacity Management
process into their design, test and production processes? What
is the most effective way to organize your Capacity Management group?
Should separate groups plan for capacity at each stage, or should it
be one central function. Join
Jim Whalen
, Service Management Director for State Farm, as he leads a
“facilitated brainstorming” session with a panel of industry
leaders in Performance and Capacity Management to address these and
other topics. Hear how companies are approaching capacity management
today, and learn the advantages and disadvantages of their current
approach.
Speaker
Biography
Jim
Whalen
is Component Manager in the IT Service group at State Farm. Jim has
over 25 years of experience in the IT industry, with over 15 years of
experience specifically focused on performance and capacity
management. Jim graduated from
Illinois
State
University
with a degree in Chemistry.
C3:
Leveraging
HP OpenView
for Predictive Performance and Capacity Management throughout the
Application Lifecycle
Doug Grumann,
Lead Engineer, Hewlett-Packard OpenView, Subram Doguparthi, Lead
Engineer, HP OpenView, Derek
Weeks
| 
