-
### 🚀 The feature, motivation and pitch
It will be great if we have an API to support evicting all KV cache from GPU memory.
By mentioning `sleep mode`, I mean, if there are some technical consi…
-
## Development Task
https://github.com/tikv/tikv/blob/ed84f5789bb34f0bc79e1fc576fac2e4a6b9dbd8/components/raftstore/src/store/entry_storage.rs#L54-L58
CachedEntries represents the raft log entries t…
-
Currently it only compacts on mutation - which leaves long expired keys in while also ignoring `compaction_interval`.
Either `compaction_interval` needs to be removed or it needs to honor the `compac…
-
Problem:
Some stacktrace is very large and it cost much memory in cache(about 2GB) and time to construct (about 4s).
In our system, 2GB memory consume can not be adopted, because 4GB can be utilized…
-
I run a Kubernetes cluster with Blobfuse CSI and we're having lots of problems from Blobfuse2 dying when too many commands are sent to it. I've managed to reproduce this with a few-liner. With any acc…
-
We're attempting to use this projects as a replacement for our homegrown memory/disk cache built around Moka and Cacache. We're seeing an issue with memory growing unbound over time, eventually leadin…
-
### Bug description
Only the page with the code blog below gives this error on average once a week.
Allowed memory size of 536870912 bytes exhausted (tried to allocate 169795472 bytes) {"exception…
-
This originated in the Memory Usage and delete_local_fixed thread but is technically a different issue so I'm opening a new thread for it and closing the old one. Initial discussion is here: https://…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Describe the new feature
When the Unleash client is called multiple times within a request, it should load the…
-
### 🚀 The feature, motivation and pitch
For prefix cache, cache hits can significantly reduce FTT. However, kv cache occupies a large amount of storage space, and the space in CPU memory and GPU vide…