When several notebooks or activities share one high concurrency Spark session, each runs in its own separate execution context, which keeps its variables apart from those of everything else in the session.
Also called read-eval-print loop.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains REPL in context, with comparison tables and the common traps.
Terms in this definition
- Share
In OpenSharing, the container you fill with tables, views, volumes, models and notebooks for recipients. You need
CREATE SHAREon the metastore to make one; add a schema and all its assets, including later ones, go with history. - Agents (classic) API
First-generation Foundry Agent Service API, based on threads, messages and runs. It is deprecated, replaced by conversations and responses, and retires on 31 March 2027.
Related terms
- FAIR scheduling
A Spark policy that shares out cluster capacity evenly among jobs that run concurrently. When notebooks share one session in Fabric's high concurrency mode, this is what splits executors between their REPL cores.
- High memory
A serverless notebook option in Public Preview giving the notebook's Python process more REPL memory at a higher cost. Spark session memory stays the same.
- REPL
Read-eval-print loop: the interactive process that executes notebook code. Serverless compute's high memory option increases this process's memory only, leaving the Spark session unchanged.