React’s rendering process is a synchronous, recursive operation. When a component throws during render, React cannot simply catch the error and continue—the component tree may be in an inconsistent state, with some components mounted and others not, some hooks already called and others not. In React 15 and earlier, this inconsistency would corrupt React’s internal…
A loading state is not a spinner. It is a contract between the application and the user: the application has accepted the request, is working on it, and will show the result as soon as it is ready. The traditional way to implement this contract is a boolean flag—const [isLoading, setIsLoading] = useState(true)—checked at the…
A React application’s JavaScript bundle grows with every feature, every third-party library, and every route. The bundle that was 200 KB at the MVP stage becomes 2 MB after a year of additions, and the user who only wants to view their profile is downloading the admin dashboard, the rich text editor, and the entire…
A database backup is not a single artifact. It is a strategy composed of different backup types, each solving a different part of the recovery problem. A full backup captures the entire database at a point in time. A differential backup captures only the pages that changed since the last full backup. A transaction log…
Sharding and read replicas are the two primary strategies for scaling a database beyond a single server, and they solve completely different problems. Read replicas create copies of the entire database that serve read-only queries, reducing the load on the primary instance for read-heavy workloads . Sharding partitions the data itself across multiple independent database…
An upsert is a single operation that inserts a row if it does not exist, or updates it if it does. The name is a contraction of “update” and “insert,” and the pattern exists because the alternative—a SELECT followed by an INSERT or UPDATE—has a race condition. Two concurrent transactions can both SELECT, both see…
Locks are the mechanism that makes concurrent transactions safe. When one transaction modifies a row, the database places a lock on it so that another transaction cannot read or write the same row until the first one finishes. This is not a bug or a performance problem—it is the foundation of the ACID model, the…
Table partitioning is the practice of splitting one large logical table into smaller physical pieces called partitions, while keeping it accessible as a single table to the application. The motivation is not raw query speed—a well-indexed table can be fast without partitioning. The motivation is manageability at scale: a table with billions of rows becomes…
HTTP status codes are the machine-readable verdict of every API call. A client reads the first digit to decide whether to parse the body, retry the request, or surface an error to the user. A monitoring system counts 5xx responses to page an on-call engineer and ignores 4xx responses because they indicate client mistakes, not…
HTTP headers are the metadata layer of a REST API. They carry information that is not part of the resource representation itself: what format the body is in, what formats the client can accept, who the client is, and how intermediaries should handle the message. For the three headers in this chapter, the distinction is…