Backend Optimization

Main Source: Various source from Google and Articles

This includes the technique of optimization in general for a backend server in 3 aspect including networks, server, and resource optimization.

Horizontal vs Vertical Scaling

Backend architecture can be scaled to handle increased traffic and load. They can be scaled in two ways:

• Horizontal: Horizontal scaling involves adding more server to distribute the load
• Vertical: Vertical scaling involves upgrading the server hardware (e.g. CPU, RAM, storage).

Network Optimization

• Minimize Requests: Combine multiple requests into a single request or use techniques like HTTP/2 multiplexing to reduce the overhead of establishing multiple connections. Minify responses file, compress or resizing image can reduce amount of data transferred over the network especially for large file.

• Content Delivery networks (CDN): Distribute static content, such as images, CSS, and JavaScript files, to a CDN. CDN is distributed servers that are strategically placed in different locations worldwide to reach user faster.

  
  Source: https://www.cloudflare.com/learning/cdn/what-is-a-cdn/

Server Optimization

• Server Caching: Implement caching mechanisms like Redis caching or reverse proxies to store frequently accessed data closer to the server.

• Load Balancing: Distribute incoming traffic across multiple servers using load balancers, this will reduce the burden of a single server.

  
  Source: https://herza.id/blog/load-balancing-pengertian-cara-kerja-jenis-metode-kelebihan-dan-kekurangan/

Resource Optimization

• Efficient Database Queries: Optimize database queries by ensuring proper indexing, avoiding unnecessary joins or subqueries, and optimizing data retrieval.

• Efficient Algorithms and Data Structures: Use algorithms and data structures that provide efficient operations for your application's use cases. For example, if you frequently search or write data, consider using hash map.

• Resource Pooling: Which is the technique of reusing resources by creating a shared repository of resources rather than creating a new resource for each request. This includes sharing database connections, network sockets, or other expensive resources.

  
  Source: https://www.thecrazyprogrammer.com/2022/01/resource-pooling-in-cloud-computing.html