implementation detail · filed under inference & serving
Round-robin routing for prefill-decode disaggregation
Uses a round-robin vLLM router with prefill-decode disaggregation enabled.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
vllm-router --policy round_robin --vllm-pd-disaggregation
usedinference servingin GLM-5.3-FlashZ.ai
Filed alongside
Other methods under inference & serving :: serving parallelism.
Attention data parallelismDeepEPPrefill-decode disaggregationEncoder-Prefill-Decode disaggregationTensor parallelism (degree 4)Tensor parallelism (degree 8)Topology-aware NVLink domain placementZero-copy fused token permutation and unpermutationData-parallel vision encodingExpert parallelismFused reduce-scatter/all-gather collectivesIdentical cache-layout pinning across prefill and decode poolsLow-precision MoE combineTensor parallelism for MoE layersTensorRT-LLM all-reduce backendToken migration for balanced expert placement