Tuning, not doping
Are your applications fit? Is your network sluggish under pressure? Regular diagnostics help detect and prevent bottlenecks at an early stage. The applications in use are constantly being consolidated. This means that the challenges IT faces are continually increasing, with growing user numbers and complexity.
By running tests throughout development you will save the time spent on searching for errors after the release. Find out how to identify faults before they appear. Analyse your application's performance and stability so that you and your IT are ready for your next marathon.
Make application performance measurable
A user's first impression of any application is vital for its future acceptance. As well as visual impressions, rapid response time behaviour, in particular, will ensure that they are satisfied over the long term. You need to be prepared for this. Using cyclical load and performance tests we can tell you how well your IT is doing.
Think about monitoring your production systems, too. An intelligent monitoring system not only measures system parameters such as CPU utilisation and current storage capacity, but also response times and their availability from the user's perspective.
For more in-depth analyses, we measure directly within your web, Java or .NET application, across all your software systems (tiers). In this way we can identify bottlenecks and suggest solutions.
If faults occur during testing or routine operations, the reason needs to be identified as rapidly as possible. For this you can use so-called root cause analyses and analyse the entire application environment, from the network through database systems to the application.
Benefits
- Substantially improve user and customer satisfaction
- Review your non-functional requirements
- Analyse the sizing of the hardware you are using
- No later optimisation costs after the release
- Identify reasons for delays, aborts and sluggish transactions
- Optimise your systems during development
Success factors
- To take performance measurements, there need to be not only functionality requirements but also quality requirements for performance, stability and upper performance limits.
- These criteria form the basis for service levels that you agree with your customers or your provider.
- Setup should be done in collaboration with the development teams or, even better, in parallel with development.
Key figures
- 750 load tests per year with a total duration of over 40,000 minutes
- Over 200 scripts generated to monitor over 190 end-to-end profiles
- 600 remote servers and over 1,000 applications monitored
Tools
- HP Loadrunner
- HP Performance Center
- HP Business Service Management
- HP Sitescope
- Apache JMeter
- Icinga / Nagios
Your opinion matters!
Did you find what you were looking for?
