Official Media Partners

1A1 : Opening the Bonnet on Application Performance

Philip Mann, Macro 4

For our American friends that's looking under the hood ---- but seriously, if your car is broken down it won't be long before you need to get the bonnet/hood open and see what is going on under there. The same is true of applications running on your computer system. You need to be able to see what is happening from the application's point of view as it executes, to be able to effectively work on improving your application(s) performance. This presentation looks at the sort of approach needed and tools available to do this; how to use them to set-up and run an effective application management function; and other problems of an organisational, managerial and political nature that can get in the way of making it all 'happen', and realising the benefits to be had from an effective application performance management function.

1A2 : Much Ado about CPU - Now with z9 Too

Martin Packer, IBM

System z processors have in recent years introduced a number of capabilities of real value to mainframe customers. These capabilities have, however, required changes in the way we think about CPU management. This presentation describes these capabilities and how to evolve your CPU management to take them into account. It is based on the author's experience of evolving his reporting to support these changes.

1A3 : Memory Matters in 2007

Martin Packer, IBM

For z/OS LPARs memory management has changed radically over the years - from both the operating system perspective and that of applications. And the pendulum has swung back and forth between focusing on Real Memory and on Virtual Memory. This presentation discusses managing both Real and Virtual Memory - from the perspectives of both the operating system and the exploiting products. The products include DB2, DFSORT, CICS, IMS, MQ and Websphere.

1A4 : Mining Gold from the RMF Data Mountain

Ivan Gelb, Gelb Information Systems

This session includes presentation of the essential RMF reports for performance management and capacity planning activities. For maximum effectiveness on the job, attendees will learn (a) important considerations for parameters affecting the data collection, (b) the minimum set of reports required to support a particular activity, (c) what are the important fields on the key reports, and (d) how to avoid some potential pitfalls. Samples of the most important and useful reports will be presented. The emphasis will be on quick techniques that help us mine the wealth of information collected by RMF.

1A5 : SAS Tips

Phil Mason, Wood Street Consultants

SAS® 9 has many new features, as well as all the old ones. This paper will try to outline some of the best new features that I have found over the last 3 years that I have been using SAS 9.

1A6 : MXG Update (Live Link)

Barry Merrill, MXG

Abstract to follow shortly

1B1 : The Capacity Management Process from A to A

Adam Grummitt, Metron

When embracing ITIL® as a discipline, Service Delivery and Capacity Management in particular should be high on the agenda for implementation to gain quick returns and meet all IT governance and compliance issues. Capacity Management covers both performance management and capacity planning and this seminar deals with it from A to A:

• Alarms & Alerting reports on current performance

• Analysis of & Advice on present & past performance

• Analytical modelling & Assurance of future performance

ITIL® V2 describes the Capacity Management process in terms of three sub-processes at resource, service and business levels. They incorporate activities such as monitoring, analysis, tuning, modelling and application sizing. Performance management entails the day-to-day addressing of the performance aspects of an IT service. Capacity planning is a strategic function used to predict IT resource requirements to meet predicted workload scenarios arising from defined business needs. It is a proactive performance assurance extension of performance management, bringing order to chaos and performance predictability to IT Service Management.

This workshop addresses all of the activities involved within the Capacity Management process. It addresses the theory, background and key concepts. It shows the nature of the main activities: measurement, analysis and prediction and reporting on all of them. It looks at the required data sources and available key metrics and the tools and techniques involved in their collection and analysis. Various statistical and analytical models available for their interpretation are introduced. The problem of client-server end-to-end response time performance measurement is addressed with reference to generic tools for modelling nodes and networks. It considers the pragmatic detail in implementing the required activities and considers their interfaces with other ITSM processes. It includes some practical exercises to underpin the techniques introduced.

The workshop is for anyone who needs an appreciation of capacity management in an ITIL® context and the issues that arise in its implementation.

• ITIL® managers who need to understand their interface with Capacity Management

• Capacity Management staff or managers liaising with outsourcers providing similar services

• Capacity Management staff requiring greater understanding of Capacity Management principles.

1B2 : The Capacity Management Process from A to A

Adam Grummitt, Metron

Please see abstract above.

1B3 : The Capacity Management Process from A to A

Adam Grummitt, Metron

Please see abstract above.

1B4 : The Roadmap for Full Lifecycle Perfomance Engineering

Adrian Johnson, Hyperformix Inc

In this paper we describe and demonstrate Performance Engineering methods and techniques that apply across the software development lifecycle. This methodology, which is based on reusable tools and repeatable processes, can be applied at any stage of the lifecycle. The session includes a case study of a retail banking application where Performance Engineering was successfully applied during design, test and production.

(Note, Chris Lee is a customer - this paper was co-authored and co-presented at USCMG in December. We don't know yet if Chris will be free to attend and present here in June, but we're working on it).

1B5 : Lifecycle Performance Management - NEVER have a Performance Problem Again in (pre) Production

Graham Parsons, Reflective Solutions

Plagued and tired of failed performance tests ? or even worse - production performance problems? Understand how your organisation can eradicate 95% of performance problems that you discover in (pre-) production ever getting into those environments. The presentation's real-World example shows how a major financial company delivered its first major project on time, under budget and with no performance problems.

1B6 : How to Select Significant Workloads in Performance Models

Paolo Cremonoesi, Polotecnico di Milano

The complexity of computer systems requires to consider the interaction of several workloads. The number of business and technical workloads potentially affecting a computer system is large, but only a limited number of them are usually required to properly model the system.

In this paper, we discuss regression-based estimates of service times required for model parametrization and we focus on the selection of significant workloads. We present an experimental comparison, using real performance logs of a distributed enterprise application, illustrating the benefits of constrained and stepwise estimations over the traditional approach based on ordinary linear regression.

1C1 : Successfully establishing Capacity Management processes

Per Bauer, TeamQuest

Among the different processes of ITIL, Capacity Management is probably the one with the most pronounced technical focus. Nevertheless, Capacity Management activities need to be designed with the overall business objectives of the organization in mind. These objectives will impact the way you choose strategies, assess risks, prioritize services and express your results. During this session, we will outline a number of critical success factors as well as best practices on how to successfully implement Capacity Management whilst keeping business objectives in mind.

1C2 : Automated detection of exception patterns

James Bouhana, Performance Int

In order to decide if systems are running according to their usual trend, it is necessary to compare against a performance baseline that defines an expected operating envelope. We describe how baselines can be derived from stores of performance data and how such baselines yield exception alerts when applied to future performance. We discuss how to use automatically derived exception alert patterns to discover breaches of normal operating modes and to assist in root-cause analysis of problems.

1C3 : MULLEN AWARD WINNER "Did something change? Using Statistical Techniques to Interpret Service and Resource Metrics"

Frank Bereznay, Kaiser Permanente MULLEN AWARD WINNER

In a perfect world, one would always know the answer to that question. Unfortunately, nobody works in a perfect world. This paper will explore statistical techniques used to look for deviations in metrics that are due to assignable causes as opposed to the period to period

variation that is normally present. Hypothesis Testing, Statistical Process Control, Multivariate Adaptive Statistical Filtering, and Analysis of Variance will be compared and contrasted. SAS code will be used to perform the analysis. Exploratory analysis techniques will be used to build populations for analysis purposes.

1C4 : Lessons Learned about Capacity Management from ITIL Training - Both as a Student and Teacher

Richard Fronhesier, Metron Technology

Abstract to follow shortly

1C5 : Building Relationships to Support Capacity Management

Sally Fitzroy and Ian Martin, Friends Provident

Having worked in the Capacity and Performance arena for many years at an assortment of companies, we joined Friends Provident during 2006 and set out to further develop the reasonably new Capacity Management processes. One of the key factors in creating a more realistic and useful capacity plan was the input received from many areas across IT and the business.

This presentation looks at the approach we took in establishing and fostering the relationships needed, the different teams we involved, how we got their buy-in, the type of information they provide and how we use it.

1C6 : Using ITIL to build a successful Capacity Management Process

Neil McMenemy, Capacitas

It may be agreed that the ITIL Capacity Management Process has some weaknesses but its major strength is in the framework that it provides, particularly in the way that Capacity Management should interface with other ITIL processes. In this slot we will look at how ITIL can be used to provide the framework for a sound Capacity Planning Process and how the ITIL self-assessment can be improved upon to enable that work to begin

1D1 : Managing Change & Capacity in the Data Centre

David Cuthbertson, Square Mile

Ensuring that the data centre delivers the reliable, computing resources needed by a business is not easy. The introduction of new technologies such as Blade servers, SANs and high performance networks is often causing system disruption as technical limits are reached. In addition, the management of change is becoming ever more crucial for consistent performance and reliability. This presentation is updated from that given at the November 2006 UKCMG forum.

1D2 : What Performance and Capacity Management People Need to Know About Finanace

Chris Molloy, IBM

Performance and capacity management (PCM) personnel need to be aware of the financial implications of their actions. IT Business models such as ITIL? have a financial management discipline included in the model which interacts with the ITIL capacity management discipline. The purpose of this paper is to describe the relationship between finance and capacity management, using ITIL as a framework for this description. The paper will extend these relationships using projects that a capacity planner may lead. It will conclude with financial mistakes made in PCM, and how one can avoid them.

1D3 : From Silos to Services

Norman Wilkinson, IBM

This session looks at how Service Level Management has traditionally been oriented towards technology and in particular, the availability of infrastructure components within technology 'silos'. This clearly has a technology focus and little overall value to the business. Today businesses are challenged with delivering core business services via IT enabled channels. These business services are often based on technically complex, composite applications that involve multitudes of touch-points within the IT infrastructure including networks, servers, mainframes, operating services and applications.

We will examine how IT can align more to business needs and objectives and transform from managing technology to managing services and re-orientate its operational culture to the service philosophy, including moving service performance measurement from just component availability to end-end service availability, management and real-time measurement of service performance against business KPI's.

This is a critical point of transition and the piece that transforms IT from a Cost Centre to a Strategic Partner for the business going forwards. We will look at where this sits in ITIL terms and the current state of maturity of most IT organisations today and a roadmap for adopting these practices in line with business expectations.

1D4 : ITIL Version 3

Steve Lawless, Purple Griffon

The presentation will cover:

• Where ITIL version has come from - what has been the driver for change

• Who has developed it and what is the new format of the Library

• What has been delivered to date and timescales for the qualifications boards and training organisations

• An overview of what has actually changed and what is still the same

• How has current 'Best Practice' in Computer measurement changed and how may this affect what we currently do

• What is still to be delivered for the ITIL Version 3 refresh - what is still in development

• Also what the future holds?.

1D5 : Method-based Troubleshooting - Rapid Resolution of IT Performance and Stability Problems

Paul Offord, Advance Seven

Method-based problem diagnosis is rarely understood and even more rarely practiced, and yet it offers an effective means to solving intermittent or long-running IT problems, cutting downtime by up to 97%. In this presentation Paul outlines an IT specific problem resolution method called RPR and describes the core principles. He moves on to describes how your company can benefit from the approach and illustrates the effectiveness of RPR with two case studies from the finance industry.

1D6 : ITIL v3 - It's impact on the Capacity Management process, function and interfaces

Rob Young, Fox IT

This presentation will look at the new Service Lifecycle approach taken within ITIL v3 and pull together the information and implications for Capacity Management. This will cover the following:

• Capacity Management within the ITIL Service Lifecycle

• The Capacity Manager role defined within ITIL v3

• Changes to the definition of interfaces to the other processes (esp. Availability, Continuity, Service Levels and Finance)

The session will end with an open discussion on the impact of these changes on the Capacity Management profession.

1E1 : Headroom - as measure of Server Capacity

Dr Prem S Sinha, Perfcap Corporation

In this session we present a metric called Headroom, which is used to quantify the remaining Capacity of a server. This metric, calculated from server measured performance data, is used to track historical change in remaining capacity. Often, in the absence of availability of business growth data a headroom trend projection can be used to predict when server capacity will be exhausted. Using PAWZ software we will show how such a methodology can be automated in support of IT resource planning.

1E2 : Windows Vista and Windows Server "Longhorn" Performance

Paul Seaton-Smith, Capacitas

Microsoft's new desktop operating system contains a number of enhanced performance monitoring tools.

1E3 : Will SOA mean the end of Capacity Management as we know it?

Bob Torz, TeamQuest

Shifting to a Service Oriented Architecture surely has implications for Capacity Management. Moving from dedicated resources to shared multi-purpose resources will render some of the assumptions of traditional Capacity Management invalid. This does however not imply that the complete process has to be redefined from the ground up. It's rather a question about modifying the traditional process and extending it by making sure some key capabilities are in place. By doing this, you can still be successful. During this session, we'll give you some hints about what those capabilities are.

1E4 : Capacity Planning for Blade Servers

Paul O'Sullivan, Positech Consulting

Abstract to follow shortly

1E5 : Rethinking Capacity Planning in a Multi-core, Multi Threaded Environment

Adrian Johnson, Hyperformix Inc

The definition of a 'CPU' is becoming more blurred as many hardware vendors have introduced multi-core and multi-threaded processor chips. Most operating systems and monitoring tools still treat each processor thread as if it were a full-blown processor. Unfortunately, this isn't quite true and the result can be significant distortion in the way that CPU utilization is reported. This paper investigates the risks of capacity planning with distorted data and how to avoid the performance traps whilst we wait for the monitoring tools to catch up.

1E6 : Where Performance Problems and Solutions are Beyond Normal Sight and Sound

Kevin Corcoran, CPT Global

In the Twilight Zone, performance problems are not always as they seem. After 20 years tuning database problems, the author has come to expect the unexpected. While 85% of all database performance problems are solved by tuning application SQL, the remainder include some very curious problems and surprising solutions. The root cause can be in the database, the operating system, the network or the SAN. On occasion more hardware helps, but sometimes it hinders. This paper describes a number of real world cases, from symptoms to diagnosis and remediation. The objective is to encourage the investigation of a broader set of causes and solutions - to enter the Twilight Zone.

2A1 : Uncaptured CPU Overheads, SRM/RMPTTOM, and evolution to the IBM System z9 EC Processor

Geoff Adams, National Australia Bank Ltd

This article details our site's experiences when implementing the new IBM System z9 EC 2094 model processors, commencing in late June 2006 .

To set the scene - a large variation in the Uncaptured CPU component was detected, with an associated fall in the capture ratio. The scale of this rise in Uncaptured was considered to be out of proportion with the increase in the measurable workloads.

After a sustained analysis effort, a mainframe industry significant resolution was unearthed towards the end of September 2006, which will be explained along with techniques and reporting methods utilised. Included will be guidance for the temporary workaround (RMPTTOM parameter), metrics as recorded at our site for different LPAR attributes, and the future enhancements to SRM as planned by IBM.

Other sites will receive the benefits of our experience if they choose to follow the revised IBM recommendations. This represents significant cost savings and is very good news for all IBM mainframe sites.

2A2 : I thought it was in Derby. A Successful Mainframe Consolidation Project

Steve Brodie, BT

A presentation, from the practical perspective, on how a mainframe consolidation project moved 67 MVS images and 60Tb of data across the country in order to meet financial targets, save data centre space and slash disaster recovery times. And no one noticed it had happened.

2A3 : z/OS Tuning Bootcamp: Back to Basics

Glenn Anderson, IBM

Whether you are new to z/OS tuning, or you are an old pro just looking for a quick refresher, you are welcome here. Glenn Anderson, who teaches z/OS performance and tuning for IBM Training, will lead you on a one hour tour of z/OS tuning hotspots, including DASD, processor storage, shared LPARs, parallel sysplex, GRS, RMF reporting, and where WLM fits in the big picture of z/OS tuning. Glenn promises this session will be way more fun than a real bootcamp.

2A4 : DB2 Data Sharing Performance for Beginners

Martin Packer, IBM

This presentation provides an introductory-level view of how to look at the DB2 Data Sharing performance numbers from both a z/OS / RMF and a DB2 perspective. Performance topics include XCF, Coupling Facility, Data Sharing Structures, the application's perspective and Structure Duplexing. Performance topics don't include other forms of Data Sharing e.g. VSAM RLS, or overly detailed descriptions.

2A5 : Reclaim your performance of DB2 V8

William R Miller, CPT Global

Many sites are moving towards DB2 V8. What are the prospective effects on performance of DB2 V8, what are the drawbacks, and what are the benefits. How soon should one move out of compatibility mode into 'entering new function mode' or 'new function mode'. This talk will concentrate on potential solutions to performance problems as DB2 V8 is introduced, with some live examples.

2A6 : WLM Caused Pain and Pleasure

Ivan Gelb, Gelb Information Systems

Workload Manager (WLM) can intentionally or unintentionally deliver pain, manifested as severe performance degradation, or pleasure, manifested as better than expected performance. For optimum and cost effective operation, attendees will learn how to insure that pain and pleasure is delivered as intended and required for the business critical applications. This session will dig into (a) the WLM service policy options by which such pain and pleasure can be controlled, (b) how to tailor these options for the results one seeks, (c) actual examples that delivered unintended results, and (d) the recommended solutions for the discussed problems. Examples will include processor, disk IO activity, and main storage caused performance degradation that could have been avoided with proper use of WLM controls.

2B1 : Analytic Techniniques to Improve Your Application Profile

Adrian Johnson, Hyperformix Inc

Abstract to follow shortly

2B2 : Making Sense out of the Spaghetti

Kobi Korsah, Wily Technology Division, CA

All too often enterprise architects create a beautiful blueprint for their SOA infrastructure only to find that the implementation is less than ideal. The reality behind the dream often consists of direct connections, legacy api's and sideways calls. Because, in truth, who can afford to deploy a new infrastructure all at once?

When undergoing a SOA sea change, Operations needs to know today's reality. The fact that the order processing application will be upgraded next week will not prevent your company from losing thousands of dollars from an outage today. There is no time to waste mapping out deployed service versions or referencing architectural blueprints. The only thing to get you out of the mess is comprehensive end-to-end SOA transaction management.

This session will define transaction management in a SOA environment. The challenges involved and the hurdles to overcome. It will discuss tools available on the market today and recommend best practices for deploying SOA components to a production environment.

2B3 : Integrating Enterprise Applications - SOA what?

David Hawkin, Severn Trent Waters

Severn Trent Water are following a strategy of Systems Oriented Architecture (SOA), whereby a single client interface provides integrated access to many of the company's key enterprise applications operating across a range of mainframe, UNIX and Windows platforms. This SOA approach significantly simplifies the job of call centre staff and at the same time yields gains in operational efficiency. Capacity Planning, supported by performance testing, has been closely involved with the development and implementation of this strategy, providing server capacity forecasts and identifying resource issues. A number of important improvements resulted in successful application roll-out without impacting to end-user performance.

2B4 : Validating functionality and performance of SOA services

Paul Murray, IBM

In order to have confidence in an SOA implementation it is important to test the services. IBM has tools that allows developers and testers to ensure the functional and performance capabilities of web services used to build an SOA implementation. This session will introduce the tools and concepts to allow users to ensure the following:

• Validate SOA system functionality and interoperability

• Ensure system performance

• Determine maximum system capacity

• Identify and resolve performance problems of SOA IT solutions

2B5 : Get the Wagons in a Circle - Running a Centralised Perfomance Testing Service

Mo Parker/Pete Pinto, Friends Provident

The highs and lows of running a centralised Performance Testing facility at Friends Provident with real-life examples of what you should and shouldn't do.

• What are the benefits and how can you best exploit them?

• What are the possible downfalls and how can you avoid a slew of arrows?

• Making sure that you're not being shot at by your own side - Building the relationships that will help improve the service.

• Still have reservations? Possible long-term implications.

2B6 : Synthetic user Testing: An Evaluation of Windows User Simulation Strategies

Neil Coleman, IBM

Understanding the application performance you are delivering to your users is critical to delivering the correct service levels. Measuring the experience of your users is more difficult with modern composite applications which remove the simple correlation between component performance and the results seen by end users.

There are many mechanisms available for measuring the user's experience of you application including log file analysis, real user session capture, synthetic testing and code instrumentation.

This presentation will focus on synthetic user simulation testing on the Windows platform and will compare the various techniques available. It will pay particular attention to the total cost of ownership of such systems as well as functionality, scalability and accuracy.

2C1 : Virtulisation - Inhibitors to Server and Storage Virtulization, and How to Mitigate Them

Chris Molloy, IBM

Virtualisation is becoming one of the leading technologies being implemented in IT environments in order to decrease costs thru increased equipment utilization. Virtualisation can occur at the server, storage, network, and application levels. The purpose of this paper is to review the benefits of server and storage virtualisation and some of the implementation technologies that are out in the industry today. The paper will then go into further detail on the inhibitors to implementing those virtualisation techniques. The paper will conclude with mitigation recommendations for those inhibitors.

2C2 : Virtualisation in Practise: Current Usage and Future Plans

Des Atkinson, Metron

This paper discusses real-world experience in the use of virtualisation technologies. For a series of client sites based in the USA and UK it describes how virtualisation is currently being used to meet business needs. It describes what technologies are being used and the kinds of applications that have been moved on to virtual architectures. It reviews the benefits and potential drawbacks in the use of virtualisation. It touches briefly on the performance and capacity issues that users have experienced. It then looks forward to describe the plans users have to expand virtualisation whether on new or established technologies.

2C3 : Application Virtualisation - the future of distributed computing?

Andy Bolton, Capacitas

Virtualisation is rapidly becoming the standard operating model for many aspects of information technology. Recently desktop applications have become the latest target of virtualisation with Microsoft's SoftGrid technology. This presentation covers the principles behind this new technology and the dramatic impact this will have on the processes, resources and infrastructure needed to manage a modern complex IT environment.

2C4 : Virtulisation Part1

Richard Talaber, VMware

This presentation draws from billions of performance samples taken on hundreds of thousands of servers from over 1000 clients around the world in order to describe what the typical infrastructure looks like today, the load demands of the typical windows server, and the importance of properly profiling application load when considering consolidation of workloads into a virtual infrastructure. Using this data we will explore strategies for implementing virtual infrastructure in your environment taking into consideration other complicating factors such as Resource Location, Service Level requirements, Security, Regulatory controls, and internal politics. Consolidation ratios and target utilization thresholds will be considered along side high availability demands, manageability, and other factors.

2C5 : Virtulisation Part 2

Richard Talaber, VMware

Please see abstract above.

2C6 : Little Boxes all the Same?.. Consolidation in the Virtual World

Bob Torz, TeamQuest

Finance departments are increasingly questioning the purchase of new hardware, so why not use the hardware that we already own and simply consolidate multiple machines into one? Virtualization now allows consolidation to be almost transparent. Simply turn several real machines into virtual ones. But now we have to size both the virtual machines and the physical machine underneath, so how can we do this quickly and accurately? Analytical modeling is one answer.

This session will explain how to use modeling to help with consolidating multiple machines from the real world into the virtual one.

2D1 : Understable Targets;The Key to a Good Service Level Agreement

Andy Packham, HCL

Agreeing Service Levels is a crucial aspect of any service implementation. However a well crafted, signed Service Level is not enough to guarantee that the service will meet the expectations of the customer. The measures and performance expectations must be communicated and understood by all the stakeholders and users of the service and by all the organisations involved in the delivery and maintenance of the service.

The Service Level Agreement provides the basis for this communication but on its own does not tell the entire story. The document often forms part of a contract so may be difficult to understand. The end-users may not be able to translate this to something workable in the real-world.

Service expectations of a new service are often high, but if the inevitable compromises made during development and implementation have not been widely communicated a gap will exist between the expectation of the user and the service provider. This expectation gap can lead to negative experience and unjust feedback directed at the service provider.

The paper will focus on how this may be avoided and the steps necessary to ensure that the expectations of the customer, the end-user and the service provider are the same, and consistent with the Service Levels.

2D2 : Managing services: ITIL, SLM, ITSM... and all that Jazz

Aurora Dell'Anno, Compuware

Attend this session to hear about implementing IT Service Management. The session will include:

• basics and buzzwords

• where are we now?

• from theory to practice: IT Service Management design implemented

• adapting Service Management best practices for IT processes

• planning for tomorrow: achieving the future

The speaker will discuss specific ways to enable and optimise IT Service Management assessment, design and planning.

2D3 : Service Level reporting, is it worth the effort?

Steve Lawless, Purple Griffon

This seminar looks at current 'Best Practice' within Service Level Management and focuses on the issues and practices surrounding Service Level Reporting. This Presentation will appeal to a mixed audience, and will give particular food for thought and insight to those considering or struggling to implement Service Level Management.

Specifically the presentation will cover;

• Why we should establish Service Level Agreements (SLAs) ?

• Current 'Best Practice' in establishing SLAs, according to ITIL

• Alternatives to SLAs

• The importance of Reporting

• What should we report?

• What should we not report?

• Best ways of reporting

2D4 : SLM Through "Software as a service"

Steve Ingall, Fox IT

Steve will look at the emerging Software as a Service market place, how this impacts on SLM and also how new SaaS products can be used to manage this situation. This will cover the benefits of SaaS; the management issues associated with using SaaS and the service management matters that need to be considered."

2D5 : Defining Availability for an IT Service

Stuart Rance, Hewlett Packard

Many customers define availability requirements for their IT Services using a very simplistic 99.9% or 5 9s figure, without any clear understanding of what these numbers might mean, how they could be measured, or how they might plan to achieve them. This often leads to dissatisfaction, with the IT organization reporting that they have met their goals even though the customer is not satisfied with the service they are receiving.

This paper discusses some of the successful approaches that have been adopted to define and measure service availability for different types of customers and services.

2D6 : Underneath the spin a practical look at service levels

Malcolm Gunn, Barclays

The setting of service levels appears easy but getting those levels to reflect the flexible requirements of the client is something that seems almost alien to Service Management.

Even the basics can be a challenge, figures are normal based around simple easy to report measures. Often we try and use the wrong figures for the wrong measure, Incident volumes to show of availability.

Remember just because its available doesn't mean it usable. Make sure the levels agreed are flexible to meet peaks and troughs in demand and are set at levels that allow the client to do their job.

2E1 : A Large Systems Guy Discovers Enterprise Workload Manager (EWLM)

Glenn Anderson, IBM

Enterprise Workload Manager (EWLM) provides a way to monitor and respond to workload processing across multiple systems in a distributed heterogeneous environment. Now there are three words that strike fear into the heart of an old MVS guy - distributed heterogeneous environment! However, EWLM is an example of mainframe technology (z/OS WLM) migrated out to distributed platforms, so that is a bit comforting. In this session Glenn Anderson, long-time WLM instructor, will share his impressions and experiences with EWLM. Just what is this EWLM thing anyway? How does EWLM fit with z/OS and WLM? Does it make sense to use both products? Why should System z people care about EWLM in the first place? Let's talk.

2E2 : Reporting/Modeling for AIX Partitioned Environments

Debbie Sheetz, BMC Software

IBM's dynamic logical partitions (LPARs) provide powerful tools for managing complex computing environments (e.g. AIX on pSeries processors). Along with the power and flexibility come new concepts, new measurements, new terminology. So how do you:

• understand what is really going on?

• report performance results to others who do not understand the new technology?

• do capacity planning for this environment?

Examples of Performance Assurance for Virtual Servers tools applied to actual customer situations will be presented, demonstrating 'best practices' for successful performance management techniques for this environment.

2E3 : AIX performance tuning

Jaqui Lynch, Circle 4

This presentation will introduce the systems administrator to some of the concepts and tools for monitoring and managing performance on UNIX systems. It will provide a guideline for how to diagnose and fix performance problems in that environment. Rules of Thumb will also be provided. Use of the new tools provided in AIX v5.3 as well as nmon will be discussed.

3 key learning Objectives

1. Know what tools are available

2. Learn how to use nmon

3. Take away a basic set of tuneables to use as a starting point

2E4 : Principles of Availability

Colin Butcher, xDelta

This session is based on many years of practical experience designing and investigating mission-critical systems. System availability is often affected by performance issues, yet the analysis of systems from an availability viewpoint is seldom part of the design or trouble-shooting process.

Business Continuity, Disaster Recovery and High Availability are important topics and are becoming more so with legislation such as Sarbanes-Oxley and with the introduction of best practice guidelines such as ITIL.

This session will discuss some of the main factors affecting system availability and some analysis techniques commonly used in other engineering disciplines. These techniques provide effective ways to think about improving overall system reliability and availability.

The session is intended to give an understanding of high availability systems and some simple ways to assess the availability aspects of a system.

2E5 : Fast track you server consolidation analysis

Walter Katz, Ideas International

2E6 : Principles of Performance

Colin Butcher, xDelta

This session is based on many years of practical experience designing and investigating mission-critical systems. Performance problems during peaks in workload are a common cause of system failure, especially in a mission-critical real-time environment. The overall performance of a system is largely determined by three main effects:

• bandwidth, which governs throughput,

• latency, which governs response time, and

• jitter (or variation of latency with respect to time), which governs the predictability of response.

This session will discuss these effects, how to observe them and ways in which they can be dealt with. The session is intended to give an understanding of the factors governing systems performance and an approach to performance trouble-shooting which can be applied in many situations, especially in those demanding high availability.

3A1 : WebSphere App Server for z/OS Ver 6 Perfmormace Tuning

Glenn Anderson, IBM

Now that you've installed WebSphere Application Server for z/OS Ver 6, how do you know that your system has been tuned for optimal performance? This session will provide performance tuning recommendations for WebSphere on z/OS, including application topology and configuration considerations, zAAP utilization, WLM settings, WebSphere tuning, JVM tuning, and z/OS system tuning . The session will also provide information about tools and techniques which can be used for performance analysis, such as RMF and the new ITCAM for Websphere.

3A2 : Paper on z/OS performance/capacity

Tony Ruberry, John Lewis

When I started as a z/OS performance analyst, I expected to be doing a 'head-down techie' job. I now know this role involves communicating with everyone from junior developers to senior management. Also, the role has changed from fire fighting to fire prevention and fire control, and this raises other challenges. In this paper I shall pass on some tips, both technical and non-technical, that I have learned to help me survive in my day-to-day job over the last six years, together with real world examples of where they have been successful

3A3 : Time is our most precious resource: Assuring Application Performance Using Automated Quality Management

Markus Bauer, TRILOGexpert

Many enterprises regard mainframe performance management, the optimization of J2EE applications in the context of service level management and end-to-end performance measurement as separate processes or even as competing processes. However, seen from an economic point of view, performance data should be correlated effectively in order to create a comprehensive performance view and an enormous added value. We present our experience and results, correlating mainframe measurements with the initiating J2EE activities, starting online transactions on the mainframe.

3B1 : Get more performance and stability by tuning your Java implementations

Graham Parsons, Reflective Solutions

Although Java was undoubtedly slow in the beginning, for a number of years global organisations have been running mission critical high-throughput, few milli-second response time applications on Java. However, out of the box, Java will not perform like this.

So how do these organisations achieve such performance and scalability. By throwing hardware at the problem? Well some do but the clever ones understand that the JVM and J2EE application servers out of the box and highly untuned.

This session will explore the opportunities for tuning the JVM and J2EE application servers, relate to real world examples stating what was performed and the results obtained - be warned, some of the figures would not be believable if it they were not fact.

3B2 : MULLEN AWARD WINNER "Java Performance Analysis 301" Part 1

Peter Johnson, Unisys

The Java Platform provides a variety of mechanisms for monitoring the performance of Java applications. There are several tools that are freely available that can be used to monitor Java applications. This session describes some of those tools, and the Java Management Extensions (JMX) technology on which those tools are built. Additionally, the paper shows how custom tools can be built using JMX.

3B3 : MULLEN AWARD WINNER"Java Performance Analysis 301" Part 2

Peter Johnson, Unisys

Please see abstract above.

3C1 : Capacity Management in the Virtual World

Matthew Auchincloss, BMC Software

Technologies such as Server Virtualisation and Service Oriented Architecture (SOA) depend upon the concept of sharing server resources in a major way. Accurate and timely capacity management of shared server resources is now more important and urgent that ever before. Matthew reviews the main Virtual technologies: AIX 5.3, HP UX, Solaris 10 Containers/Zones, VMWare, as well as Hyperthreading and Simultaneous Multithreading. Matthew discusses the challenges these technologies present to the Capacity Manager, and reviews ITIL based approaches to allow these technologies to be exploited in a cost effective way.

3C2 : Planning/Sizing for Virtualization (capacity planning for consolidating machines)

Jaqui Lynch, Cirlce 4

In this session Jaqui will discuss what virtualization is in the midrange and will go over how that works in the System P servers. She will also discuss some of the issues around server consolidation using virtualisation and will highlight a few traps for young players. Issues relating to planning and sizing will also be covered.

3 key learning Objectives:

• Know the traps to avoid

• Understand the virtualization concepts

• Understand the various workload manager options

3C3 : Do CPUs Count? Understanding Resource Utilisation on Virtualised Systems

Debbie Sheetz, BMC Software

Distributed systems vendors have developed virtualization technologies that allow multiple application servers to run within a single hardware complex. The benefits include (a) more flexibility in administration and deployment, (b) increased overall utilization of the underlying resources, and (c) corresponding significant reductions in the cost of the IT infrastructure. However, successful management of virtualized hardware environments requires enhancements to analyzing performance statistics compared to what was done for dedicated server systems. Understanding what is real vs. virtual is critical in reducing risks to service delivery. This presentation provides a technical overview of vendor technology and interpretation of performance measurements for UNIX Virtual Servers, VMware, and Microsoft Virtual Server.

3D1 : Achieving IT Service Excellence Through Effective Performance Management

Richard Seery, SAS

The goal of Performance Management is to actively and proactively manage an organization to assure that it achieves pre-determined levels of performance. It refers to the use of Performance Measurement information. IT organizations typically monitor data provided by a range of tools that represent technical performance points such as configuration management. Management of technical performance is an important role but IT management is also responsible for meeting their organization's highest priorities and business needs with available resources. ITIL V3 is based on the Service Lifecycle with the aim of aligning Business and IT Service Strategy. This paper discusses how Performance Management along with ITIL V3 will assist an IT organizations endeavors in achieving IT Service Excellence.

3D2 : Making Profits More Profitbale Using Metrics

Jon Ward, Severn Group

Abstract to follow shortly

3D3 : Open Analytics - Exploiting Performance and Avaialbility Models to Reduce Stakeholder Dissonance

Dr Chris Tofts, Hewlett Packard

The consensus view on IT Management is that it should achieve alignment with the business. What is it the IT is supposed to align with? A business is a collection of cooperating (?) stakeholders with different views on the role of the organisations IT. The problem of IT alignment is identifying where the compromises between these groups are made.

Model based workshops are effective means of identifying the underlying business dynamics of import for these stakeholder groups, and used the models to illustrate their impact upon each other, and the IT systems of their various positions.

3WKS1 : Advanced WLM

Glenn Anderson, IBM

Today your z/OS system is filled with transactions and server address spaces of all types. Remote DB2 queries, Stored Procedures , WebSphere Application Server, CICS, IMS, WebSphere MQ, Unix Daemons, etc., etc. How does the Workload Manager (WLM) deal with all these different kinds of work? It uses a number of WLM services - enclaves, application environments, execution delay monitoring services along with a combination of response time and velocity goals. This session will cover these advanced WLM services in detail along with proper use of classification rules and RMF reporting. A basic knowledge of z/OS WLM is a prerequisite for this session.

3WKS2 : Performance Engineering

Adrian Johnson, Hyperformix

Abstract to follow shortly

3WKS3 : Virtualisation and its Impact on Capacity & Performance Management

Paul Seaton-Smith, Capacitas

Virtualisation products, such as Microsoft Virtual Server or VMWare Server, are increasingly becoming the standard in many data centres and test beds. These allow multiple guest operating systems to run under a single host operating system on a server, thereby appearing as separate discrete servers. While virtualisation has been around on the mainframe platform for many decades it has only been used relatively recently on distributed Operating Systems (OS) such as Unix variants and Microsoft Windows Server. On Unix platforms virtualisation has become popular due to vendor-led initiatives such as containers in Solaris. However Microsoft Windows Server has only more recently been taken seriously with the introduction of VMWare, so much so that Microsoft have had to play catch-up by launching their own product Virtual Server with inferior functionality. Whilst the mainframe platform has addressed these issues for many decades, for less-mature platforms, such as Windows Server, the increasing use of virtualisation brings its own problems with monitoring, measuring, analysing and controlling workload on the platform. This seminar is designed to review some of these issues faced by companies investigating virtualisation and provide some suggestions as to risk mitigation and effective capacity management in a virtualised environment.

This seminar is designed to:

• Describe current virtualisation trends and the market impact

• Describe the main virtualisation technologies on Windows Server and Solaris

• Describe the challenges of capacity planning a virtualised environment

• Describe some solutions to capacity management in a virtualised environment

©2008 UK Computer Measurement Group