Snowflake Architecture
| Course Details | ||||||||
|---|---|---|---|---|---|---|---|---|
|
||||||||
| Languages | ||||||||
|
||||||||
| Outcomes | ||||||||
|
||||||||
| Price & Terms | ||||||||
|
||||||||
| Technical | ||||||||
|
||||||||
| Specification | ||||||||
|
||||||||
Learn how Snowflake’s compute, storage and services layers work, size virtual warehouses, load data, tune clustering and protect data with Time Travel.
Snowflake keeps compute, storage and cloud services in three separate layers, and the course traces how that split shapes daily work on the platform. You size virtual warehouses for different workloads, load structured and semi-structured data, see how micro-partitions and clustering keys let Snowflake prune what a query reads, and recover or copy data with Time Travel and zero-copy clones. The aim is a foundational picture of how the platform is built and why it works the way it does.
What You Will Achieve
- Choose an edition and a warehouse configuration that suit a workload, and spot a misconfigured warehouse
- Set up a database structure, grant the right privileges and load structured and semi-structured data into tables
- Judge whether a column makes a good clustering key and pick the right table or view type for a job
- Query past data with Time Travel, test a zero-copy clone and match Cortex functions to tasks such as PDF extraction
Program Overview
Snowflake Architecture and the Object Hierarchy
How the three layers divide the work, how editions differ, and how objects, access and parameters are organized in a Snowflake account.
- Compute, storage and cloud services layers
- Snowsight, notebooks and Query History for failed queries
- Object, access control and parameter hierarchies
- Filtering with session variables
Virtual Warehouses and Loading Data
Building the Snowy Peak database, granting privileges, tuning warehouses for different workloads and loading its data into tables.
- Database objects and privileges
- Warehouse types, sizing, scaling and concurrency
- Loading structured and semi-structured data
Storage Internals, AI/ML Tools and Data Protection
How data is physically organized, which table and view types fit which needs, the AI/ML tool set, and protecting and sharing data.
- Micro-partitions, partition pruning and reclustering
- Table types, materialized views and unstructured files with pre-signed URLs
- Snowpark, Streamlit, Cortex, Snowflake ML and Snowflake CoCo
- Time Travel, zero-copy cloning and Marketplace sharing
Certificate Awarded (Paid)
Job Outlook
These skills are used wherever teams run their data platform on Snowflake: setting up warehouses and access, loading data for analysis, tuning how tables are stored and sharing data with other groups.
Good for: Analysts and data engineers starting on Snowflake who want to see how warehouses, storage and access control fit together before building on them.
Keep in mind: The AI/ML tools share one chapter with Time Travel, cloning and sharing, so Snowpark, Cortex, Streamlit and CoCo each get a short look next to the architecture material.
Bottom line: A grounding in how Snowflake’s layers, warehouses and micro-partitions work, practiced on the Snowy Peak case rather than on abstract diagrams.
Frequently Asked Questions
What is Snowy Peak in the exercises?
Snowy Peak is the case the hands-on work is built around. You set up its database structure, grant privileges, load its user data and query its avalanche event data while working through warehouses and data loading.
Which Snowflake AI and ML tools come up?
The final chapter matches Snowpark, Streamlit, Notebooks and Snowflake Cortex to their uses, compares Cortex with Snowflake ML, picks a Cortex function for PDF extraction, and uses Snowflake CoCo to query app sessions and find PII with its built-in skills.
How are unstructured files handled?
Alongside table types and views, the storage chapter covers working with unstructured files in Snowflake, including generating a pre-signed URL and a use case that calls for one.
Will I use Snowflake Notebooks during the lessons?
Yes. Early on you run a query in a Snowflake Notebook as one of the ways of working with the platform, and the AI/ML chapter returns to notebooks to look at their properties.
User Reviews
Write your review
You must be logged in to post a review.
Reviews can only be posted by registered learners. Log in or create a free account, then you will return here to post.
There are no reviews yet.