1

The basics of IT performance

• The four pillars of performance : availability, robustness, response time, scalability.
• From availability management to high availability : law of 9, MTBF, MTTR.
• How to evaluate the availability of an application ?
• Contracting process of services levels (SLA).
• What about Java, .NET and PHP platforms ?
• Comparison of levels of performance with «old platforms » : mainframe, AS 400.
• Are Open-Source solutions compatible with high levels of performance ?
• What are the main problems one comes across on new IT architectures ?

2

Include performance in project management

• From a curative management of performance to a preventive approach.
• When to make prototypes ? On which parameters ?
• Carry out technical tests in the development phase : methods and tools.
• estimate the costs of integration phases and the cost of transitions to the production phase.
• The role of ITIL in the production process.
• Applications designed for production. Facilitate configuration of applications.
• Improve communication between applications and the production teams' tools
• Architectural and development best practices : The impact of Java and .Net virtual machines on performances.
• The new programming constraints.
• Patterns and good coding practices
• Profiling tools.

3

Hardware infrastructures and performances

• Servers and processors : The impact of multi-heart CPUs ; The new mainframes (Unix and Windows)
• Storage : The impact of RAID technologies on performance ; What are SAN and NAS approaches ? ; Distributed file systems
• Networks : Virtual IP addresses ; Solutions for load-distribution on networks.
• Data-Centers : Choice of location. Power supply, ventilation, wiring. WAN links.
• Server virtualisation : Operating principles ; Solutions : VMWare, Virtual Server , XEN ; Virtualisation and performance

4

Software infrastructure and performances

• Clustering, principles of « load balancing » and « session fail-over ».
• Active/Passive type clusters. Active /Active type clusters. Hybrid clusters.
• The notion of session affinity.
• Application server clusters
• Data-base clusters.
• Pros and cons of clustering.
• Grids and on demand solutions : Principles, advantages. Offers (IBM, Oracle...)

5

Technical acceptance and transition to production phase

• Technical tests : Specify technical tests : examples of deliverable ; How to estimate costs ? ; High availability valid
• Metrology of applications : Definition, goals and advantages. Methodology ; Load injection tools (Mercury, Compuware, Si
• Deployments management : Preparing platforms and applications
• Development
• Production : Managing incidents

6

Monitoring and supervision

• Monitoring : Life cycle of a monitoring solution ; Key documents to produce and maintain ; Information gathering on avai
• Application supervision : Defining indicators and their different states ; Defining thresholds, rules and alarms ; Compo
• SNMP standard : Principles, MIB and basic orders ; SNMP support in Java 5 and Windows
• JMX standard : Principles ; JVM monitoring with MXBeans ; Mbeans specific development ; Overview of available tools arou
• WMI standard : Principles. Reading and writing data. Designing and developing a WMI provider, tools.
• APM consoles and tools : Quest Foglight. Quest PerformaSure. BMC Patrol. IBM Tivoli. HP Openview. Microsoft Operation Ma

7

Anticipate needs

• Managing capacity : What is capacity planning ?
• Approach of capacity planning
• Experience feedback.