Our experience with performance certification of enterprise application shows that root cause of majority of performance bottleneck lie in Bad Architecture. Generally architecture of application are designed keeping in mind the immediate need of business and fail to reckon the long term scalability requirement of the user load on the system
The purpose of the Software Architecture Review service is to provide an advanced technical analysis of your application with practical documented recommendations on how to amend and finesse the architecture to ensure a highly scalable, reliable and extensible applications and a repeatable process for your organization.
Through our team of experienced performance consultants, who have extensive experience of designing and implementing complex enterprise level applications across verticals like CRM, Insurance and Travel, we assess your architecture for scalability, reliability and extensibility
We can help you by providing the following services
- Application Architectural review for Performance
- Network Architecture review for Performance
Time required for the review process depends on the size and complexity of the application and the number of issues covered. Large applications nearing the end of development with overt technical issues can extend the review process to two to three weeks.
During the review, the consultant will need to have access to your application. This usually includes:
- Source code
- An executable application from within a development environment.
- The partitioning scheme for the application.
- A running, fully deployed version of the application
The consultant may also require access to developers, architects, testers and other people involved in the development and deployment process.
Some of the topics that the consultant will discuss with you include:
- Network: examines such factors as speed of the physical network, objects traversing the network, messaging etc.This requires participation by your networking specialists, and may also require specialised equipment such as network sniffers, packet analysers, etc.
- Load balancing: is the right number of load-balanced replicates being used, is load balancing not being used where it should, or vice-versa?
- Reliability: explores how failover is being taken advantage of in the application, and what additional considerations there might be.
- Database: while this is not typically a comprehensive performance review, the consultant may delve into details such as database indexing, cache sizing, O/S kernel parameters, and other low-level details.This is usually based on specific client requirements, will require participation by your database administrators, and should be clearly communicated to the consultant prior to the engagement.
- Deployment considerations: examines what should be the ideal deployment architecture of the application