How Databricks serves queries
Pause
FOR CONTEXT
How query-ready data enables fast answers
Fresh-data path
Newly
written
Fresh
data
continuously
Query-ready data
Ordered
raw data
Ordered pre-
aggregated
Query engine
Query
compute
Fast
answers
continuously
Drill-downs
Interactive aggs
Databricks
Fresh-data path
Zerobus
Ingest
Arrow Flight
streams
NBBO
quotes
1M rows/s
Liquid-clustered
Delta table
Incremental MV
Ordered
raw data
Ordered pre-
aggregated
Async
serverless
clustering
Async
serverless
MV refresh
Serverless SQL
warehouse
X-Small
≈16 worker vCPUs
Query
results
May scan
unordered data
May read
stale data