APM

The performance of a web-based application depends not only on the application server itself but also on various components and interfaces in the application delivery ecosystem. Bottlenecks, which can occur at one or more nodes of the application delivery ecosystem (e.g., the end user device through which users can access the application, the network that connects the application servers with user interfaces, the backend databases, and so on), have an adverse impact on application performance. Application Performance Management (APM) service helps users monitor, detect, and diagnose complex application performance problems to maintain an expected level of service.

NIC’s APM service is an agent-based solution for managing the performance, availability, and user experience of applications and is offered as Software-as-a-Service from NIC National Cloud. The service is based on a client-server model where monitoring agents shall be installed on the various components of the application delivery ecosystem (that need to be monitored) to collect performance data. The data shall be sent to the APM server (which is managed by the NIC APM team), which shall analyse the collected data and present actionable insights through a dashboard with drill-down capabilities.

APM Service Deployment Architecture

The APM measures application performance by using these two sets of metrics. While the first set of performance metrics defines the performance experienced by end users of the application, the second set of performance metrics measures the computational resources used by the application for the load (volume of transactions processed by the application, e.g., transactions per second (TPS), requests per second, pages per second), indicating whether there is adequate capacity to support the load, as well as possible locations of a performance bottleneck. The measurement of these quantities establishes an empirical performance baseline for the application. The baseline can then be used to detect changes in performance. Changes in performance can be correlated with external events and subsequently used to predict future changes in application performance.

• End-user experience monitoring: Depicts end-user response time trend, i.e., the time taken from request initiation to receipt of the result set. Helps in viewing the overall end-user pertinent performance metrics.
  
  
• Application discovery and modelling: APM automatically discovers the application topology, which depicts how components are interacting with each other. The discovered flow map is dynamic in nature and changes with changes in application deployment/nature of traffic. The APM automatically baselines different metrics collected from the application environment and makes it easy for the assignment of health rules, alerts, and appropriate policies. Dynamic baselining is "learning what is normal from past behaviour."
  
  
• User-defined transaction profiling: The APM solution automatically discovers business transactions and constructs a topology map of how traffic flows within the application. This helps users manage the application based on accurate user behaviour.
  
  
• Application component monitoring: The platform provides correlated and contextual visibility into the infrastructure and other correlated metrics. The infra and/or database metrics can also be charted on a custom dashboard for continuous monitoring.
  
  
• Reporting & Application data analytics: The Geo Dashboard helps users identify the application experience from different locations in a region. It can be cities within a country or state or also private centres within a single city. You can drill down into each location and investigate the requests originating from that location.
  
  
• APM Consultation and Support from NIC: The NIC APM team shall provide technical consultation to users regarding the APM service.