Training for Snowflake SnowPro Core Certification [COF-C03]

Explore Snowflake features and architecture, including storage, compute, and services layers, objects and their container hierarchy, billing, and connections like Snowpark and Snowflake Scripting.

Discover Snowflake as a cloud-native data platform offered as a service, uniting data warehouse and data lakehouse capabilities with features like Snowpipe, Openflow, and Cortex for diverse workloads.

Snowflake built a multi-cluster shared data architecture for the cloud, decoupling storage, compute, and management services into separate layers to enable independent scaling and workload isolation.

Snowflake's storage layer uses the cloud blob store (example amazon s3) with high availability and durability, storing data in a centralized, encrypted columnar format with micro-partitions for efficient sql queries.

Explore the query processing (compute) layer in Snowflake, featuring virtual warehouses that scale out compute, pause and resume, isolate workloads, and ensure strict ACID processing via the transaction manager.

Explore the Snowflake services layer, including authentication and access control, infrastructure management, transaction and metadata management, query parsing and optimization, and security, to understand global coordination across accounts.

Explore Snowflake editions—Standard, Enterprise, Business Critical, and Virtual Private Snowflake—and learn how each edition affects compute and data storage costs, security, governance, and service levels.

Explore the Snowflake object model, including databases, schemas, tables, streams, stages, and pipes as securable objects and account level objects within an organization.

Explore how Snowflake organizations manage multiple accounts with organization accounts, using ORGADMIN and GLOBALORGADMIN to create, rename, or view accounts, and how databases and schemas form a namespace.

Explore the four basic Snowflake table types—permanent, temporary, transient, external—and their time travel and fail-safe retention, plus edition differences and external tables’ read-only status.

Learn about standard and materialized views in Snowflake. Standard views store a query and fetch data from source tables; materialized views maintain a pre-computed dataset with storage and compute costs.

Discover how Snowflake workspaces replace worksheets, migrate by importing SQL files, and run SQL across environments with familiar context, settings, and navigation changes.

Demonstrate a hands-on Snowflake workflow in Snowsight, working with databases, schemas, tables, and views, covering permanent, temporary, transient tables and external tables.

Open table formats enable database-like guarantees on data lakes using Apache Iceberg within Snowflake. Understand how metadata, snapshots, and a catalog model data and ensure interoperability across engines.

Explore how Snowflake parameters—account, session, and object—shape behavior with alter commands, precedence rules, and practical examples like periodic_data_rekeying.

Configure account, session, and object parameters in Snowflake—set date formats and defaults, enable Workspaces as the default sql editor, and explore data retention time across database, schema, and table.

Explore how stored procedures in Snowflake bundle SQL statements and procedural logic, focusing on JavaScript-based procedures, signatures, return types, privileges, and the call syntax, and compare them with UDFs.

Explore UDFs, external functions, and stored procedures in Snowflake with hands-on exercises that create, call, and manage SQL and JavaScript code, including API integrations and cross-language use.

Explore how sequences in Snowflake generate sequential and globally unique numbers using NEXTVAL, with configurable start and increment, and apply them to table inserts or default column values.

Learn to orchestrate data processing with tasks and streams in Snowflake, using scheduled sql statements or procedures, and connect tasks in a directed acyclic graph to capture and apply changes.

Learn how Snowflake dynamic tables declaratively keep data up to date from a query, enabling simple, multi-step pipelines without streams or tasks, with automatic refresh, configuration, and monitoring.

Learn to issue SQL statements from the command line with SnowSQL, enabling automation for CI/CD pipelines, interactive sessions, and data loading against a Snowflake account.

Learn to author and execute Snowflake SQL directly in Visual Studio Code with the Snowflake extension, manage connections and profiles, and view results and query history.

Explore Snowflake connectivity through connectors and drivers, including Python and Spark, plus Kafka and JDBC/ODBC, and learn how Partner Connect links BI, data integration, security, and ML tools.

Explore Snowflake scripting to implement procedural logic in sql, including declare blocks, begin and end transactions, variables, loops, branching, and exception handling, with anonymous blocks and stored procedures examples.

Compare on-demand and pre-purchased capacity pricing and learn how Snowflake credits drive compute costs. Understand storage and data transfer billing, cloud services adjustments, and serverless features.

Execute sql and python Snowpark in Snowflake notebooks, using pre-installed numpy, matplotlib, and pandas for data analysis. Schedule recurring notebook tasks and manage compute with virtual warehouses.

Explore Snowflake cortex and AISQL, Copilot, and Document AI, learning how llms power SQL functions, SQL generation, and PDF data extraction for analytics.

Fine-tune a smaller model on sentiment labels from a large model to reduce costs and improve accuracy; leverage Cortex Search with embeddings for semantic queries in Snowflake.

Master building interactive web apps with Streamlit, a Python framework that simplifies widgets, data visualization with pandas, and deploying Snowflake Streamlit apps with Snowpark.

Explore account access and security basics, including RBAC, discretionary access control, roles and privileges, authentication methods such as MFA and federated or key pair authentication, and data encryption.

Explore Snowflake access control frameworks—RBAC, discretionary, and user-based—role ownership, privilege grants, and object hierarchies to enforce least privilege across securable objects.

Explore Snowflake roles and privileges, including role hierarchy, inheritance from child to parent, and creating custom roles to enforce least privilege.

Explore Snowflake security privileges that govern access to securable objects, including global, account, schema, and schema-object privileges, with grant and revoke and future grants.

Explore Snowflake access control by creating and managing roles, grants, and privilege inheritance, including custom roles, future grants, and object ownership.

Learn how Snowflake authenticates users with a username and password, create users with the create user privilege, and enforce strong password rules with optional must change password settings.

Discover how multi-factor authentication with Duo Security strengthens Snowflake login using push notifications, calls, or passcodes, with automatic per-user enrollment and easy setup via the Snowflake interface.

Explore federated authentication with Snowflake, enabling secure single sign-on through IdPs like Okta or ADFS, including the SAML_IDENTITY_PROVIDER properties and IdP initiated vs Snowflake initiated workflows.

Generate a 2048-bit private key pair and assign the public key to a Snowflake user. Enable key pair authentication via connectors and rotate with two active keys to avoid downtime.

Create network rules and network policies to allow or block inbound IPv4 addresses, then apply the policy to an account or user, understanding precedence and optional bypass settings.

Snowflake automatically encrypts data at rest with AES-256 and in transit with TLS 1.2, enabling end-to-end security across internal and external stages. Learn about key rotation and tri-secret secure.

Discover how column-level security uses dynamic data masking to protect sensitive data, applying masking policies to columns and enabling authorized users to see plaintext while others see masked results.

Apply row level security with row access policies to restrict rows by current role. Compare them to masking policies, anchor to a column, and use boolean policy bodies.

Set up and test column and row level security with masking and mapping table policies, using system admin, analyst, and masking admin roles to control data access.

Snowflake secure views restrict access to underlying tables and view definitions to authorized owners, and bypass some query optimizations, potentially slowing performance. They are discouraged unless strict security is required.

Explore how the Snowflake account usage views and the information schema provide historical usage, credits, and metadata across accounts, with latency and access considerations for reporting.

Tag Snowflake objects with a business unit to organize and govern data, enabling masking policy integration and inheritance across tables, schemas, and columns.

Snowflake alerts are schema-level watches that evaluate a condition, execute an action, and trigger on a schedule or new rows, with serverless or warehouse compute options.

Explore virtual warehouses in Snowflake, including configurations, billing, state, and properties. Cover resource monitors, multi-cluster warehouses, and sql tuning, caching, materialized views, automatic clustering, and the search optimization service.

A virtual warehouse is a named abstraction for a massively parallel processing compute cluster, comprised of worker nodes in the cloud that power Snowflake queries.

Learn how virtual warehouses transition between started, suspended, and resizing states, how auto suspend, auto resume, and initially suspend affect billing, cache, and readiness to process queries for control.

Learn how Snowflake virtual warehouses manage credits, states like started and suspended, and workload isolation for concurrent queries. See hands-on creation, sizing, auto-suspend and auto-resume, and on-the-fly configuration.

Explore how Snowflake uses t-shirt sizing to match virtual warehouse resources, and learn how workload complexity, data loading, and per-second billing affect performance and cost.

Learn to set Snowflake resource monitors to cap credit usage with daily, weekly, or never quotas and trigger notify or suspend actions at account or warehouse levels.

Track Snowflake compute usage and credits via the usage tab, account usage views, and information schema; then configure resource monitors with quotas and triggers to manage warehouse suspensions and notifications.

Explore how snowpark optimized warehouses increase memory and customize CPU architecture, compare with standard warehouses, and enable Python and machine learning workloads in Snowflake.

Enable the query acceleration service on a virtual warehouse to burst serverless compute for complex queries. Speed up execution, improve concurrency, and monitor eligibility and scale factor to balance cost.

Analyze Snowflake query performance using query history, query profile, and account usage views to identify long-running queries, bottlenecks, and optimization opportunities.

Learn practical SQL tuning for Snowflake: optimize queries with early where filters, proper joins on unique columns, mindful grouping and ordering, and avoiding spills by managing memory and cardinality.

Snowflake caching uses the metadata cache, the results cache, and the local disk cache (SSD) to speed queries, with a 24-hour results cache and an LRU warehouse cache.

Explore how the metadata cache and results cache speed up queries, suspend and resume virtual warehouses, and observe local vs remote storage effects on performance and costs in Snowflake.

Explore materialized views, which store pre-computed query results and refresh automatically to balance freshness and performance. Learn limitations, costs, and use cases, including external tables.

Explore clustering as a way to describe a column's value distribution and how Snowflake uses micro-partitions and pruning to speed queries, using clustering metadata such as overlaps and depth.

Implement Snowflake's automatic clustering by defining clustering keys on tables or materialized views to reorder data in micro partitions, improving query performance while considering serverless costs and data changes.

Demonstrates how well-clustered keys boost pruning and speed up queries on large Snowflake tables, by examining clustering metadata, overlap, and depth with hands-on exercises.

Enable search optimization on a table to speed up lookups and equality searches using = or in, for integer, numeric, date, time, timestamp, varchar, and binary values via access path.

Master data loading and unloading in Snowflake, including insert and copy into table, serverless features, and exporting with copy into location and get commands, plus semi-structured data and functions.

Explore simple data loading techniques in Snowflake, including insert statements and loading via the UI from CSV, with mapping, file formats, and a hands-on walkthrough.

Explore how Snowflake stages manage data loading and unloading, comparing internal and external stages, including user, table, and named internal stages, and using the put command and storage integrations.

Explore the Snowflake stages—user, table, and named internal stages—and learn to view, upload, and manage files, including external stages and storage integration with AWS S3.

Learn how to use the copy into <table> bulk loading command to move data from internal or external stages into Snowflake tables, with format options, transformations, and validation.

Snowflake uses file_format options on stages or COPY INTO statements to identify file types like CSV, JSON, and XML, and to handle headers, compression, and encoding during loading.

Learn how to bulk load data from a stage into a Snowflake table using copy into, configure file formats, and handle errors with validation mode and on error continue.

Snowpipe automates near real-time data ingestion by loading staged files into a target table via a pipe, triggered by cloud messaging or a REST call, with two modes and serverless.

Discover how Snowpipe Streaming delivers data in as little as five seconds by streaming rows via channels into a target table, bypassing the staging layer.

Use the Snowpipe Streaming Python SDK to stream transactions into Snowflake via key pair authentication, registering public key and configuring the Transactions table and TRANSACTIONS_SIMULATOR channel for real time delivery.

Unload data from Snowflake using copy into location and get commands, exporting to internal or external stages or cloud storage in formats like CSV, TSV, JSON, and Parquet with partitioning.

Explore semi-structured data in Snowflake, including JSON, XML, arrays, objects, and the variant type, and learn how Snowflake extends SQL to load and query these formats.

Load and unload semi-structured data by uploading files to a stage and copying into table storage with Snowpipe, using json or parquet formats and variant columns.

Learn to query semi-structured data in a Snowflake variant column using dot notation, bracket notation, and the GET function to access nested elements and arrays.

Learn how Snowflake's flatten table function and lateral flatten explode semi-structured data, turning arrays like skills into table rows with path, sequence, and value details.

Learn to load and query semi-structured json data in Snowflake using variant, flatten, and lateral join, with etl and automatic schema detection, through a hands-on session.

Explore data transformations in Snowflake, examining structured and unstructured data, Snowflake functions, estimation functions, unstructured file functions, rest api support, directory tables, and table sampling.

Explore Snowflake’s function families—scaler, aggregate, window, table, and system—and learn data generation with unique IDs, min/max, plus table and system functions for query management and Snowpipe status.

Explore how Snowflake uses estimation functions to approximate cardinality, similarity, frequency, and percentile with HyperLogLog, MINHASH, Approx_Top_K, and T digest for faster insights.

Explore table sampling to read a subset of rows using fraction-based methods (sample or table sample) with row or Bernoulli probabilities, seed, fixed-size row counts, plus block or system sampling.

Explore how Snowflake handles unstructured data by generating access URLs from internal or external stages using three file functions, including time-limited and permanent options, with required privileges and secure sharing.

Directory tables provide a queryable view of files in internal and external stages, exposing permanent file_url, size, hash, and last_modified, with manual or automated refresh to keep metadata current.

Explore the Snowflake file support REST API to download data files from internal or external stages using get /api/files, with scoped URLs or file URLs and access permissions.

Discover secure data sharing, data exchange, and the data marketplace, then inspect the storage layer with micro partitions and data protection features like time travel, replication, and fail-safe cloning.

Explore Snowflake’s storage layer: centralized scalable blob storage (s3/azure blob/cloud storage), auto micro-partitions, data reorganization into compressed and encrypted columnar format, supporting structured and semi-structured data accessed via SQL.

Ingest data via copy into table, Snowflake automatically partitions into micro-partitions stored in blob storage, using columnar storage, min/max metadata, and micro-partition pruning for efficient queries.

Learn how Snowflake uses time travel and fail-safe to safeguard data, with configurable retention, undrop, at, and before, plus cloning to recover and analyze past table states.

Configure time travel retention across account, database, schema, and tables using data retention time in days, explore undrop, show history, and use at and before to restore data.

Explore how Snowflake cloning creates metadata-only copies of tables, databases, and more, using zero-copy cloning within the retention period, with recursive cloning and privilege considerations.

Master zero copy cloning in Snowflake by creating table and database clones as metadata pointers, with recursive cloning, point-in-time cloning within retention, and exceptions for external tables and stages.

Snowflake replication enables cross-account database replication within an organization for business continuity and disaster recovery, using a read-only secondary refreshed from the primary; some objects and privileges do not replicate.

Learn how Snowflake calculates storage billing based on daily on-disk data for tables and internal stages, including time travel, fail-safe, and retention impacts, with regional per-terabyte pricing and example costs.

Learn how secure data sharing lets a Snowflake data provider grant read-only access to objects for a consumer without transferring data, enabling cross-account queries and pay for access model.

Explore secure data sharing in Snowflake through private sharing, inbound and outbound shares, and reader accounts; learn to create shares, grant privileges, and access data across accounts.

Explore Snowflake's data marketplace in Snowsight, where data providers publish standard or personalized listings for data consumers, enabling rapid access, customized requests, and potential charges.

Learn how to create a private data exchange in Snowsight, and invite providers and consumers to share data. Manage access, approve listings, and monitor listing requests.