Technologies/Trino/trino.memory.pool.free_bytes
TrinoTrinoMetric

trino.memory.pool.free_bytes

Free memory in pool
Dimensions:None
Available on:Native (1)DatadogDatadog (1)
Interface Metrics (2)
Native
Available free memory in the memory pool
Dimensions:None
DatadogDatadog
Free distributed memory available across the cluster
Dimensions:None
Sources
trino.memory.MemoryPool.FreeBytestrino.io
trino.memory.free_distributed_bytesdocs.datadoghq.com

Technical Annotations (18)

Configuration Parameters (11)
task.low-memory-killer.policyrecommended: total-reservation-on-blocked-nodes
Enables automatic query termination before memory exhaustion with TASK retry policy
retry-policyrecommended: TASK
Enables task-level retries requiring memory management
query.low-memory-killer.policyrecommended: total-reservation-on-blocked-nodes
default policy kills highest-memory query on OOM nodes
query.low-memory-killer.delayrecommended: 5m
Grace period before killing OOM queries
sink.max-buffer-sizerecommended: 64MB or 5120MB
Worker-side buffer for data being written to downstream stages
exchange.max-buffer-sizerecommended: 64MB or 5120MB
Buffer size for data exchange between stages
exchange.deduplication-buffer-sizerecommended: 64MB or 5120MB
Buffer for deduplicating exchange data
driver.max-page-partitioning-buffer-sizerecommended: 5120MB
Buffer for partitioning pages in drivers
sink.max-broadcast-buffer-sizerecommended: 512MB
Buffer for broadcast sink operations
exchange.max-response-sizerecommended: 32MB
Maximum response size for exchange operations
exchange.max-error-durationrecommended: 5m
How long to tolerate exchange errors before failing query
Technical References (7)
trino_events.trino_queriescomponentpeak_memory_bytescomponentOOMconceptleast-waste policyconceptretry-policyconfigurationexchangecomponentsinkcomponent
Related Insights (11)
Memory exhaustion requires manual query kills with TASK retry policycritical
▸
TASK retry policy without memory killer requires manual interventioncritical
▸
High memory queries risk OOM and cluster instabilitycritical
▸
Insufficient resources cause query failurescritical
▸
Memory pressure blocks worker nodeswarning
▸
Low memory triggers query killer on out-of-memory nodescritical
▸
Low memory triggers task killer on out-of-memory nodescritical
▸
Memory exhaustion triggers query and task terminationcritical
▸
Insufficient buffer sizes prevent large dataset operationswarning
▸
Queries killed by low memory killer during OOM conditionscritical
▸
Tasks killed by low memory killer with inefficient waste handlingcritical
▸