Struct MockEngineArgsBuilder 

pub struct MockEngineArgsBuilder { /* private fields */ }

Builder for MockEngineArgs.

Implementations

impl MockEngineArgsBuilder

pub fn num_gpu_blocks(self, value: usize) -> Self

pub fn block_size(self, value: usize) -> Self

pub fn max_num_seqs(self, value: Option<usize>) -> Self

pub fn max_num_batched_tokens(self, value: Option<usize>) -> Self

pub fn enable_prefix_caching(self, value: bool) -> Self

pub fn enable_chunked_prefill(self, value: bool) -> Self

pub fn watermark(self, value: f64) -> Self

pub fn speedup_ratio(self, value: f64) -> Self

pub fn dp_size(self, value: u32) -> Self

pub fn startup_time(self, value: Option<f64>) -> Self

Optional startup time in seconds to simulate engine initialization delay

pub fn worker_type(self, value: WorkerType) -> Self

Worker type for disaggregated serving (Aggregated, Prefill, or Decode)

pub fn perf_model(self, value: Arc<PerfModel>) -> Self

Performance model for timing predictions (not serialized, loaded from planner_profile_data)

pub fn enable_local_indexer(self, value: bool) -> Self

Enable worker-local KV indexer for tracking this worker’s own KV cache state

pub fn bootstrap_port(self, value: Option<u16>) -> Self

Bootstrap port for disaggregated serving rendezvous. Prefill workers listen on this port; decode workers connect to it. If None, bootstrap rendezvous is disabled.

pub fn kv_bytes_per_token(self, value: Option<usize>) -> Self

KV cache bytes per token, auto-computed from model config by Python CLI. Formula: num_layers * 2 * num_kv_heads * head_dim * dtype_bytes

pub fn kv_transfer_bandwidth(self, value: Option<f64>) -> Self

KV cache transfer bandwidth in GB/s for disaggregated serving latency simulation. Default: 64.0 (inter-node InfiniBand). Set to 0 to disable KV transfer delay. For intra-node NVLink, typical value is ~450.

pub fn reasoning(self, value: Option<ReasoningConfig>) -> Self

Reasoning/thinking token configuration. When set, the mocker wraps output in thinking boundary tokens.

pub fn zmq_kv_events_port(self, value: Option<u16>) -> Self

ZMQ port for publishing KV events in vLLM’s native wire format. When set, the scheduler publishes to a ZMQ PUB socket instead of directly to NATS. A KvEventPublisher relay subscribes to this socket and forwards events to NATS.

pub fn build(self) -> Result<MockEngineArgs, MockEngineArgsBuilderError>

Builds a new MockEngineArgs.

Errors

If a required field has not been initialized.

Trait Implementations

impl Default for MockEngineArgsBuilder

fn default() -> Self

Returns the “default value” for a type.