QuestionQ120

VerifiedEnsuring solution and operations excellence

The application reliability team at your company has added a debug feature to its backend service that sends all server events to Google Cloud Storage for later analysis. The event records are at least 50 KB and at most 15 MB, and they are expected to peak at 3,000 events per second. You want to minimize data loss.

Which process should you implement?

Explanation

Storing each event as a separate compressed object limits the loss from any individual failed upload. Including metadata in the object body ensures it is persisted with the event in the same write operation. Randomized object-name prefixes distribute high-volume writes more evenly; Cloud Storage recommends avoiding sequential names at high request rates because they can constrain performance.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!