.NET Core custom eCommerce Solution - Digital Signature Leader

We implemented repository pattern architecture in .NET Core along with GraphQL as query language for APIs. For the front-end we used React.js. We used the following components to achieve the client's requirement.

• .NET Core Token-Based Authentication with JWT : The implementation of token-based authentication using JSON Web Tokens (JWT) adds an extra layer of security to the system. By passing the token in a cookie instead of the header, it makes it harder for unauthorized users to copy or manipulate the token, enhancing the security of user sessions.
• .Net Core API Throttling and Request Limiting : The inclusion of API throttling and request limiting based on custom rulesets allows for better control and security. Different countries may have specific limitations, and users' plans can also affect the rate at which requests are allowed. This feature helps prevent brute force attacks and improves overall security by limiting excessive requests.
• .Net Core GraphQL Integration : Instead of traditional RESTful APIs, the implementation of GraphQL as a query language for the API offers advantages such as resolving over-fetching and under-fetching of data. GraphQL allows clients to specify the exact data they need, reducing unnecessary data transfers and improving overall productivity and performance.
• .Net Core & Redis Caching : As the ECommerce application is hosted on multiple servers, we decided to use Redis distributed cache. The Caching has reduced database calls and significantly improves performance.
• .NET Core Logging with Azure Monitor and Kusto Query Language : Azure Monitor is utilized for logging purposes. We used Kusto Query Language (KQL) to fetch specific types of logs, facilitating tasks such as fraud detection (e.g., identifying unauthorized use of coupon codes for a 100% discount). Created Grafana dashboard by using KQL query. Provided a dashboard to Analyze error rates and trigger events such as sending emails, SMS notifications, or PagerDuty alerts are achieved through these logging capabilities.
• .NET Core New Relic Integration : New Relic's SDK is implemented into the application to monitor application and system health. It allows tracking key metrics like CPU utilization and raises alerts if certain thresholds are exceeded. Monitoring application health with New Relic helps identify potential issues and ensures proactive management of system performance.
• .NET Core Telemetry logging :Added specific telemetry logging into the application which helps to provide more insights of our specific feature or entire application. We used Microsoft Insights for telemetry logging.
• .NET Core Hosted Services & Background Jobs : The tasks which do not require user input or interaction are managed using .NET Core hosted services or background jobs. Fraud detection, Sending emails and text notifications, Data Validations & Migration are few tasks which will be managed by .Net Core hosted services or background jobs.
• .NET Core & Feature Flags Approach : Every new feature is managed by a specific flag. .NET core developers followed the practice to verify the feature flag at all places during development. The Feature flag approach helped us to enable-disable any functionality without redeployment.
• .NET Core Unit Testing and Code Coverage : We made sure that Unit Test classes are added for each functionalities. We follow 80% code coverage rules to make sure code works in various scenarios.
• GIT & GITHub Actions : Sprint based development with each sprint having a new branch. Created various GITHub actions to perform unit tests, code coverage, database scripts and sanity checks
• .NET Core & Jenkins CI / CD : Implemented CI / CD using Jenkins to deploy code base to multiple servers. As a part of the process it runs the unit tests if it passes, it generates the build image and runs on all production servers.
• Load Balancing : Load balancing is implemented on each environment, distributing incoming traffic among multiple servers. Different load balancing methodologies such as Source IP Hash and Resource-Based (SDN Adaptive) load balancing are employed to optimize the distribution of traffic based on various factors.
• .NET Core integration with Optimizely (A / B Testing) : We used Optimizely to implement A/B testing. Every new feature introduced will be rolled out to a certain percentage of the user base. When work fine gradually increased to 100 percent or rollback if required.