scaling capabilities of neon?
wanted to know what how scale can I get with neon.
is it ready to handle 170 millions users. did you guys do any tests on number of conncurrent queries and limits and other scaling issues.
I am scared cuz I saw in a video saying you guys off load cold data to s3 is it reliable. me as a customer knowing it's going on s3 and if that data need to be access it will first do s3 request to get into the disk then process seems slow to me at scale or am I wrong . I have no experience in database just my normal brain saying this. Please let me know about the scaling capibilites, specificly 170 milliosn monthly active users will neon be able to handle this kind of scale? I don't even have 1 user but I am qurious about the scaling capabilities of neon.
Thank you
1 Reply
variable-lime•2y ago
Dear @Alvee, using a pooled connection, you can expect up to 10 000 concurrent connections out-of-the-box.
This can be achieved simply by adding "-pooler" to your connection string.
Please have a look at our documentation for further details, here is the relevant page: https://neon.tech/docs/connect/connection-pooling
Should you need more concurrent connections, it's possible to offload the read queries to a read-replica, which means that you could have one read/write endpoint handling up to 10k concurrent connections and additional replicas handling 10k concurrent connections each (See: https://neon.tech/docs/introduction/read-replicas).
If you need more than 10k concurrent writes, please, raise a support case providing as much details as possible and I would be happy to dig into our internal monitoring and bring this point to our engineering team attention. We may be able to fine tune settings here and there to help you achieve your objective 🙂
Neon
Connection pooling - Neon Docs
Neon uses PgBouncer to offer support for connection pooling, enabling up to 10,000 concurrent connections. PgBouncer is a lightweight connection pooler for Postgres. This topic describes Neon's defaul...