specific method · filed under inference & serving
Sub-NUMA partitioning with per-VM NUMA binding
A node-level resource-management method that uses hardware-supported sub-NUMA partitions and binds each worker VM to a NUMA domain.
- 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.
At the node level, we use hardware-supported sub-NUMA partitioning and bind each worker VM to an individual NUMA domain.
usedsoftware implementationin DeepSeek Elastic Compute (DSec)DeepSeek
Filed alongside
Other methods under inference & serving :: inference scheduling.
Asynchronous schedulingNUMA binding of workers to GPU-local CPU socketsComposite-key sorting for topology-aware GPU rank assignmentContinuous batchingCross-group pinning of cache-hit blocksDual-cluster consistent-hash failover for cache affinityEnvoy-based proxy with custom orchestratorExacto routingGreedy feasible-rank placement by remaining capacityHost-side scheduling optimizationIncreasing batch size for inferenceLatency-sensitive execution class with priority isolationMax sequences tuningPer-node hard admission constraintPrefix-cache-aware session affinity schedulingRegistering NVLink domains as Ray custom resourcesRequest-class resource-budget admission controlRuntime-signal-based rollout concurrency auto-throttlingWorkload-aware routed-expert GEMM scheduling