implementation detail · filed under inference & serving
Identical cache-layout pinning across prefill and decode pools
Pins KDA convolution-state and KV-cache layouts identically across the prefill and decode pools.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- core
How sources treat it
One count per evidence span, weakest treatment to strongest.
core 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
The KDA conv-state and KV-cache layouts must be pinned identically on both pools
coreinference 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 collectivesLow-precision MoE combineRound-robin routing for prefill-decode disaggregationTensor parallelism for MoE layersTensorRT-LLM all-reduce backendToken migration for balanced expert placement