Enterprise Application Modelling
Enterprise wide applications are increasingly commonplace, and can often comprise a number of smaller applications, across a multitude of platforms, in a variety of locations. This complexity can make performance slow and capacity requirements costly. Capacitas can model complex, multi-tier, multi-location applications to ensure that the performance requirements can be met at a reasonable cost.Capacitas can develop models that predict the impact of a new application on the existing enterprise infrastructure. These predictive models can be used by either Capacitas or the client to determine the impact of increasing traffic volumes or new applications on the infrastructure utilisation and hence the impact of response times. The models can then be used as part of the client's own regular capacity management process or as part of the Capacitas Capacity Planning & Management Competency Framework.
Capacitas uses a range of techniques to build predictive models; including linear regression models, analytical models, queuing network models and sophisticated discrete event simulation models. The choice of techniques depends on budget, time, application complexity and service level agreements.
Capacitas can use these models to predict:
- The new application response time
- The impact on the existing application response times and throughput
- Performance bottlenecks on the existing infrastructure at the different layers (e.g. presentation, application and data layers)
Key Benefits
- Typically modelling may be conducted without the need for expensive modelling tools, allowing mitigation of targeted performance and capacity risks with a relatively low cost solution
- Capacitas will model the application in its entirety, including all technology layers (UNIX, Windows, LAN, WAN, wireless LAN, etc)
- Clients will benefit from Capacitas's extensive experience in modelling common applications, e.g. MS Exchange, File & Print Services, Web Services (MS IIS, Apache, etc) and J2EE applications