NTD 2023

Tendences in IT industry empower companies to scale, move their systems and build entire infrastructure in cloud. This is driven by various economical and technological factors that are difficult to complain with. Such transformations require complex integrations, rebuilding of applications and usage of virtual infrastructure that naturally initiates the need of performance, security and other non-functional testing. Higher complexity architecture and infrastructure requires more detailed and complete testing of mentioned aspects.
There are specifics between application deployments in cloud and on-premise that significantly impacts purpose and approach of performance testing:

• Residency of applications in virtual machines, containers and other forms provided by cloud platforms (i.e. lambda – serverless computing in AWS).
• Automated scaling of resources.
• Embedded redundancy and partial reliability of concrete services.
This requires enhanced performance testing approaches and models of how such activities/services can be performed. Performance testing approach for cloud applications considers next cloud specific aspects:
• Flexible configuration of cloud environment and resource scaling capabilities leads to enhanced set of tested parameters, application and infrastructure tunning towards optimized usage.
• Standard set of observability tools enables tracking of detailed statistics on different levels.
• Data management capabilities (i.e. using snapshots) support multiple executions.
• Deployment of test tools in cloud allows to emulate close to real life traffic.

Meanwhile performance testing delivery model depends significantly on projects – complex and long-term projects require dedicated performance engineers or even teams, but smaller size transformations can follow particular patterns. This leads to industrialized model of performance testing delivery by optimizing reusable assets, sharing of engineering resources, process standardization and brings cost optimization for clients. Such service can be deployed for individual client or can serve multiple clients in parallel.

Key takeaways:
This presentation provides a fresh look on specifics of applications in cloud and elaborates on enhanced approaches for various types of performance testing and application monitoring supported by case study of enterprise level media platform testing in cloud. Audience will learn specifics of performance testing in cloud, modern approaches and tools and benefits of running these practices during 3+ years long continuous project.