SnowPro COF-C03 glossary of Snowflake AI Data Cloud, governance, loading, performance, and collaboration terms.
Use this glossary when Snowflake storage, compute, governance, loading, and collaboration terms start to blur together. Keep it beside the cheat sheet and resources instead of using it as a substitute for practical review.
| Term | Short meaning | Why it matters on COF-C03 |
|---|---|---|
| Virtual warehouse | Snowflake compute cluster used for query and data-processing work | Core compute and cost term |
| Cloud services | Snowflake control-layer services that coordinate metadata, optimization, and management behavior | Important architecture boundary |
| Micro-partition | Internal Snowflake storage unit used for pruning and performance behavior | Core storage and performance term |
| Stage | Snowflake location used for loading or unloading files | Core ingestion term |
| File format | Snowflake object that defines how staged files are parsed | Common loading boundary |
| Snowpipe | Continuous data-ingestion service for loading files into tables | Commonly confused with COPY INTO and integrations |
| Storage integration | Snowflake object that enables secure access to external cloud storage | Common loading and external-stage term |
| Time Travel | Feature for querying or restoring historical object state within a retention window | Core recovery term |
| Fail-safe | Additional recovery window managed by Snowflake after Time Travel expires | Commonly confused with Time Travel |
| Clone | Zero-copy snapshot of an object at a point in time | Commonly confused with recovery and sharing |
| Secure share | Snowflake data-sharing mechanism that exposes data without copying it | Core collaboration term |
| Reader account | Snowflake-managed account pattern that lets a provider share data with someone who does not already have a Snowflake account | Collaboration delivery term |
| Listing | Publishable Snowflake data product or offering for consumer discovery and controlled access | Marketplace and collaboration term |
| Clustering key | User-defined key that can improve pruning for some table access patterns | Common performance concept |
| Result cache | Cached query result reuse when conditions allow | Common performance and cost term |
| Query profile | Execution details view used to understand query behavior | High-value troubleshooting term |
| Network policy | Access control mechanism that restricts connection sources | Connectivity and security term |
| Resource monitor | Snowflake control used to notify, suspend, or otherwise react to warehouse-credit consumption thresholds | Monitoring and cost term |
| Masking policy | Governance feature that can hide or transform sensitive data values for certain roles | Governance term |
| Notebook | Interactive Snowflake workspace for code, SQL, and analysis | Current COF-C03 feature-awareness term |
| Cortex | Snowflake AI capability family exposed inside the platform | Current COF-C03 feature-awareness term |
| Apache Iceberg table | Open table-format concept relevant to Snowflake interoperability and current-version feature awareness | Current COF-C03 feature-awareness term |
VARIANT |
Snowflake data type used to store semi-structured data such as JSON | Core data-type term |
| Pair | Keep this distinction clear |
|---|---|
| warehouse vs database | compute resource versus stored data container |
| file format vs stage | parsing rules versus file location reference |
| connector or driver vs storage integration | client connectivity versus secure external-storage access |
| Time Travel vs Fail-safe | self-service historical access versus additional managed recovery window |
| secure share vs clone | cross-account data access without copy versus point-in-time duplicate object |
| result cache vs warehouse scaling | reused output versus more compute capacity |
| masking policy vs role grant | data obfuscation rule versus permission assignment |
| notebook vs worksheet or CLI | interactive development surface versus simpler query or command surface |
| Cortex vs generic SQL feature | built-in AI capability versus ordinary query syntax |
| Iceberg table vs native table | open-format interoperability versus ordinary Snowflake-managed table behavior |
| role hierarchy vs object hierarchy | privilege inheritance boundary versus data-object organization boundary |
Use these when the answer choices all sound “Snowflake-ish” but only one matches the real platform role:
| Cluster of terms | Keep this clean mental split |
|---|---|
file format vs stage vs COPY INTO |
parse rules vs file location vs load action |
| Clone vs Time Travel vs secure share | duplicate object vs historical recovery vs cross-account collaboration |
| warehouse vs micro-partition pruning vs result cache | compute scaling vs storage-layout selectivity vs reuse of prior output |
| role hierarchy vs object hierarchy vs governance policy | authorization boundary vs data-object boundary vs protection rule |
| notebook vs Cortex vs Iceberg | development surface vs AI capability vs open-table-format interoperability |
| Topic family | Best page to revisit |
|---|---|
| architecture and performance pickers | Cheat Sheet |
| current Snowflake docs and exam references | Resources |
| pacing and review order | Study Plan |
| overall exam framing | Guide root |
Do not use this glossary as a substitute for platform reasoning. Use it when: