Juncheng Gu
Juncheng Gu
In Linux kernel 4.x, swap-in (read) request is allowed to batch multiple pages. Multiple pages can be in a single swap-in request. Infiniswap was initially developed on kernel 3.13 which...
This PR adds TPU support in `NixlConnector` (#17751) for P/D disaggregated serving. The high-level idea is to use a buffer in host memory as the kv transfer buffer. The kv...
### Your current environment The output of python collect_env.py ```text CPU(s): 224 On-line CPU(s) list: 0-223 Vendor ID: AuthenticAMD Model name: AMD EPYC 7B13 CPU family: 25 Model: 1 Thread(s)...