Sharding is how most of us scale Postgres. It works, however, it is not without its challenges. What key should we shard on? How big will my shards need to be? How will sharding affect performance? What application-level mods will need to happen? What happens if we need to re-shard the database?
Automating the sharding process seems right. We should be able to define a shard and let the db take care of the underlying segmentation and movement of data.
CockroachDB is wire-compatible with PostgreSQL and automates the sharding process. If you want to scale, you simply spin up a node, point it at the cluster and the database takes care of rebalancing and moving data. You can scale on the fly and even reshard while in production without downtime or loss of data.
In this webinar, we will speak through how CockroachDB delivers this effortless scale, covering: Primary keys in Cockroach - the “key” to sharding RAFT, range partitions and the architecture that delivers scale Automating data placement for latency and resilience Re-shard in production without downtime All while speaking SQL