Best practices for improving performance #33659
-
Hello, please help me which methods are most effective in improving the performance of a vector database.
there is a collection of ~25 million records.
The table takes up about 27 GB of memory space We' re calculating hash of key to select partition to read/write from software side (not There are 128 partitions in total. QPS:
Milvus deployed in k8s (not standalone). At the moment, it has been possible to achieve an indicator of 4k QPS of reading operations. k8s config:
At the moment, the collection is loaded as 8/4 in-memory replicas. There are no changes in the performance here.
The rest of the architecture: RootCoord, QueryCoord, etc - under one replica Avg CPU usage in QueryNodes like 50% May you please give any advice about tuning perfomance in this situation I will be grateful for any answer. Thank you. |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 8 replies
-
Memory usage is 27GB. Some suggestions:
Change the replica_number to 4 and observe the QPS.
|
Beta Was this translation helpful? Give feedback.
-
Hello! I need some advice to horizontal milvus scaling. BTW I got 10k QPS with this setup. Also using eventually consistencyLevel. And there is an interesting moment. Perfomance requirements has changed and we need 25k QPS. Expectations: 10k QPS -> 30k QPS Also increased resources for proxyNodes and queryCoords, but it like useless. |
Beta Was this translation helpful? Give feedback.
Memory usage is 27GB.
Query nodes RAM = 64 * 4GB = 256GB, CPU cores = 64 * 2 = 128
Some suggestions: