May 2019
Intermediate to advanced
442 pages
11h 36m
English
A sharding strategy, whether vertical or horizontal, while necessary in certain situations, shouldn't be taken lightly. The complexity in managing multiple configurations for multiple Prometheus instances quickly adds up, making your life more difficult if you didn't plan the required automation accordingly. Things such as external labels, external URLs, user consoles, and scrape jobs can be programmatically set to reduce the operational toil for the team maintaining the shards.
Having a global view also becomes a problem as each Prometheus instance will have its own subset of the data. This can make dashboarding harder as the location of the data may not be immediately clear (for example, when using multiple ...
Read now
Unlock full access