v6d
v6d copied to clipboard
vineyard (v6d): an in-memory immutable data manager. (Project under CNCF, TAG-Storage)
Created from https://github.com/v6d-io/v6d/issues/1873#issuecomment-2071294061
Hi, I found it quite troublesome to get a C++ client all the way through compiling the entire project and trying to understand the raw C++ APIs in the doc,...
Describe your problem --------------------- Also need to bump the relevant components such as golangci-lint. Fixes https://github.com/v6d-io/v6d/security/dependabot/84 Additional context ------------------ https://github.com/v6d-io/v6d/pull/1823 https://github.com/v6d-io/v6d/pull/1824
Describe your problem --------------------- As tittled. Currently disable vineyard memory usage checks of llm test provisionally.(Fix in the future)
Describe your problem --------------------- Currently, we will fetch and get a block object if user request a cache block(update or query) and create a builder from this block object. Then...
Describe your problem --------------------- During the integration with llama.cpp, we find the import_kv_cache_buffer and export_kv_cache_buffer show bad performance as follows.  The Import time includes `query vineyard` + `import_kv_cache_buffer`. The...
Describe your problem --------------------- As tittled. Temporarily close.
Describe your problem --------------------- Since we have the python API of llm kv cache, we can integrate it with [vllm](https://github.com/vllm-project/vllm).
Describe your problem --------------------- Currently, the perfect hash map used by the vertex map has problems with the string type. Replace BBHash with pthash to align with libgrape-lite.