Replies: 4 comments 3 replies
-
Yes, we would like to to add sorted secondary index support to FASTER, once the general secondary indexing framework (work in progress) is merged in. And yes, you can keep sorted list of keys and continue to use FASTER for lookups. |
Beta Was this translation helpful? Give feedback.
-
I have written an experimental B+ Tree backed by FasterKV with promising performance (at least it is faster than SortedSet at the moment.) I don't want to put more effort into it if native Range Queries are coming soon. I want to learn more about the ongoing feature for secondary indexes and Range Queries. Could you please share here relevant PRs, Discussions, and Documents? If someone shares the deadline or estimation for these ongoing features, that would be pleasant. Thank you. |
Beta Was this translation helpful? Give feedback.
-
@cmadh I have not open-sourced that because it was just experimental work. Anyway, I am attaching the code here for those interested. After my experiments and search on the topic, I decided to write an LSM tree with classical Lock-Based implementation. I am going to open-source that soon. |
Beta Was this translation helpful? Give feedback.
-
@cmadh I have finally completed the LSM Tree for .NET and have made it open-source with an MIT license. |
Beta Was this translation helpful? Give feedback.
-
I understand that Faster uses HashTable internally. That means the keys are not ordered.
Is there a desire to add a sorted data structure to Faster in the future?
Would it be good practice if I keep a sorted list of all keys in my application and continue to use Faster for lookups?
I would appreciate any other workaround suggestion.
Beta Was this translation helpful? Give feedback.
All reactions