memory usage DiskANN vs HNSW #35318
Unanswered
rere950303
asked this question in
Q&A and General discussion
Replies: 2 comments 5 replies
-
For DiskANN, 8GB memory is probably good for 1.5-2m 3072 dimension data. |
Beta Was this translation helpful? Give feedback.
5 replies
-
all the parameter is related to recall. you can always use default value if you don't have too much idea about how to config except for search_list |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
When using DiskANN with vector index, how much less memory is used compared to when using HNSW? The vector dimension is 3072.
Beta Was this translation helpful? Give feedback.
All reactions