MindTree Performance Engineering Services

Performance Engineering starts at the design stage with performance objective definition. However, conventionally companies have been carrying out Performance Engineering at very the end of the software development life cycle.

There have been multiple instances, where the products and applications have met the functional requirements but have failed to meet the performance and end-user expectations. This has often led to increase in development cost, problem identification, and resolution time as well as poor customer satisfaction. Overlooking these factors can prove to be detrimental in today's highly competitive markets.

MindTree follows a consultative approach by comprehending your business concerns and developing a solution with a quantifiable ROI.

Three key aspects of MindTree's Performance Engineering approach are:

• Defining performance - How fast is fast enough? Comprehending the customer's perception of performance and re-evaluating it from the view point of the end-user, the competition, and existing industry standards
• Building performance - Tuning and building performance based on performance objectives and analysis
• Measuring performance - Tracking relevant metrics to measure your product's/application's performance against the objectives

MindTree leverages its Performance Engineering services to develop software products and applications that perform optimally in normal as well as extreme load conditions, thus reducing the number of failures related to performance and availability. MindTree offers Performance Engineering services across a wide range of verticals and applications based on client server, Web technologies, Web services, and ERP.

MindTest Performance Engineering Methodology

MindTree follows a 4 step methodology for executing end-to-end performance engineering. MindTree defines a customized model encompassing all or a few of these steps based on customer requirements, performance objectives, and product / application's readiness for performance.

Key Testing Types and Metrics

Testing Types	Load Reliability Scalability Failover Recovery Stress Testing
Metrics Tracked	Client side and server side monitoring Garbage collection data # of thread contention events heap memory usage CPU / RAM utilization over time Application/web-server/database server slow down/freeze/crash # Page loads per unit time over time # managed code exceptions thrown per second

MindTree Performance Engineering Differentiators

• Full-cycle performance engineering services
• Proprietary frameworks to accelerate the process and optimize testing
• Extensive expertise in databases, high volume transaction systems, and related technologies
• On-demand performance analysts and architects to improve product performance
• On-demand capacity planning and sizing assistance
• Deep expertise in proprietary, open source (e.g. JMeter), and third-party tools (Load Runner, VSTS)
• Repeatability of tests and results

Performance Testing: Process

The activities involved in conducting performance testing for varied applications on Client server, Web technologies, Web services, and ERP are detailed below:

Experience

MindTree has provided cutting-edge performance and load testing services for a leading international airline, and helped identify and fix critical memory issues for call management software from the world’s leading customer care solution provider. We have also load-tested multi-tiered credit limit applications for a leading multinational Bank.

MindTree is tool agnostic, and will identify and work with solutions that are best suited for your unique requirements. Our experience with various tools within application testing covers Load Runner, e-Load, Web LOAD, Silk Performer, Rational Performance Tester, QA Load, JMeter, and The Grinder.

Load Runner Execution Process

MindTree’s Load Runner Execution process starts with generating virtual user scripts that drive applications as clients. Load Runner Controller then manages the load tests. Read >>

Sample Metrics

MindTree has always adopted a strong metrics-driven approach. Cross-result and transaction response time graphs are just some of the metrics used in performance testing. Read >>