
Description
Guides Qdrant query latency optimization. Use when someone asks 'search is slow', 'how to reduce latency', 'p99 is too high', 'tail latency', 'single query too slow', 'how to make search faster', or 'latency spikes'.
SKILL.md
Scaling for Query Latency
Latency of a single query is determined by the slowest component in the query execution path. It is sometimes correlated with throughput, but not always — throughput and latency are opposite tuning directions.
Low latency optimization is aimed at utilising maximum resource saturation for a single query, while throughput optimization is aimed at minimizing per-query resource usage to allow more parallel queries.
Performance Tuning for Lower Latency
- Increase segment count to match CPU cores (
default_segment_number: 16) Minimizing latency - Keep quantized vectors and HNSW in RAM (
always_ram=true) - Reduce
hnsw_efat query time (trade recall for speed) Search params - Use local NVMe, avoid network-attached storage
Memory Pressure and Latency
RAM is the most critical resource for latency. If working set exceeds available RAM, OS cache eviction causes severe, sustained latency degradation.
- Vertical scale RAM first. Critical if working set >80%.
- Use quantization: scalar (4x reduction) or binary (16x reduction) Quantization
- Move payload indexes to disk if filtering is infrequent On-disk payload index
- Set
optimizer_cpu_budgetto limit background optimization CPUs - Schedule indexing: set high
indexing_thresholdduring peak hours
Vertical Scaling for Latency
More RAM and faster CPU directly reduce latency. See Vertical Scaling for node sizing guidelines.
What NOT to Do
- Do not expect to optimize latency and throughput simultaneously on the same node
- Do not use few large segments for latency-sensitive workloads (each segment takes longer to search)
- Do not run at >90% RAM (cache eviction causes severe latency degradation that can last days)
- Do not ignore optimizer status during performance debugging
- Do not scale down RAM without load testing (cache eviction causes days-long latency incidents)
More skills from the skills repository
View all 30 skillsqdrant-advisor
diagnose and troubleshoot Qdrant deployments
Jul 16DatabaseDebuggingOperationsQdrantqdrant-clients-sdk
integrate Qdrant client SDKs
Jul 16API DevelopmentDatabaseQdrantSDKqdrant-deployment-options
select Qdrant deployment options
Jul 19DatabaseDeploymentInfrastructureQdrantqdrant-edge
build applications with Qdrant Edge
Jul 16DatabaseEdgeQdrantSearchqdrant-horizontal-scaling
guide Qdrant horizontal scaling decisions
Jul 19ArchitectureDatabaseQdrantScalingqdrant-hybrid-search
implement hybrid search in Qdrant
Jul 16AI InfrastructureDatabaseQdrantSearch
More from Qdrant
View publisherqdrant-hybrid-search-combining
combine hybrid search scores in Qdrant
skills
Jul 16DatabaseQdrantSearchqdrant-hybrid-search-prefetches
configure hybrid search with prefetch queries
skills
Jul 16API DevelopmentDatabaseQdrantSearchqdrant-indexing-performance-optimization
optimize Qdrant indexing and data ingestion
skills
Jul 19DatabaseDebuggingPerformanceQdrantqdrant-memory-usage-optimization
optimize Qdrant memory usage
skills
Jul 16DebuggingPerformanceQdrantqdrant-model-migration
migrate embedding models in Qdrant
skills
Jul 16DatabaseMigrationPerformanceQdrant