What is Snowflake - Visual Explanation

Process Flow - What is Snowflake

User wants to store data

↓

User uploads data to Snowflake

↓

Snowflake stores data in cloud storage

↓

User runs queries on data

↓

Snowflake processes queries using compute resources

↓

Results returned to user

Snowflake lets users upload data to the cloud, stores it safely, and runs queries quickly using separate compute power.

Execution Sample

Snowflake

CREATE TABLE users (id INT, name STRING);
INSERT INTO users VALUES (1, 'Alice');
SELECT * FROM users;

Creates a table, adds one user, then retrieves all users.

Process Table

Step	Action	Evaluation	Result
1	Create table 'users'	Table does not exist	Table 'users' created with columns id and name
2	Insert row (1, 'Alice')	Table 'users' exists	Row added to 'users'
3	Select all from 'users'	Table 'users' has 1 row	Returns [(1, 'Alice')]
4	End of commands	No more commands	Execution stops

💡 All commands executed successfully

Status Tracker

Variable	Start	After Step 1	After Step 2	After Step 3
users table	does not exist	exists empty	exists with 1 row	exists with 1 row

Key Moments - 2 Insights

Why does Snowflake separate storage and compute?

Snowflake stores data separately from compute so queries can run fast without affecting storage. See execution_table steps 3 and 4 where query runs independently.

Is data stored inside Snowflake or outside?

Data is stored in cloud storage outside Snowflake's compute layer, but Snowflake manages it. This is shown in the concept_flow step where data is uploaded and stored in cloud storage.

Visual Quiz - 3 Questions

Look at the execution_table, what is the result after step 2?

A Returns [(1, 'Alice')]

B Table 'users' created with columns id and name

C Row added to 'users'

D Execution stops

💡 Hint

Check the 'Result' column for step 2 in execution_table

At which step does Snowflake return query results?

A Step 3

B Step 2

C Step 1

D Step 4

💡 Hint

Look at the 'Action' and 'Result' columns in execution_table

If we add more rows before step 3, how does it affect the result at step 3?

A Query will fail because table is too big

B More rows will be returned in the query result

C No change, only one row is returned

D Table will be deleted

💡 Hint

Refer to variable_tracker showing table rows after insertions

Concept Snapshot

Snowflake is a cloud data platform.
It separates storage and compute for speed and scalability.
Users upload data to cloud storage.
Compute resources run queries independently.
This allows fast, flexible data analysis.

Full Transcript

Snowflake is a cloud-based data platform that stores data separately from compute resources. Users upload data, which Snowflake stores in cloud storage. When users run queries, Snowflake uses compute resources to process them without affecting storage. This separation allows fast and scalable data handling. For example, creating a table, inserting data, and querying it shows how Snowflake manages data and compute independently.

Practice

1. What is Snowflake primarily used for?

easy

A. Managing network security

B. Creating mobile applications

C. Storing and analyzing data in the cloud

D. Designing websites

Solution

1. Step 1: Understand Snowflake's main purpose

   Snowflake is a cloud service designed to store and analyze data easily.

2. Step 2: Compare options with Snowflake's use

   Options B, C, and D relate to other fields like app development, security, and web design, not Snowflake.

3. Final Answer:

   Storing and analyzing data in the cloud -> Option C

4. Quick Check:

   Snowflake = Data storage and analysis [OK]

Hint: Snowflake = cloud data storage and analysis [OK]

Common Mistakes:

• Confusing Snowflake with app or web development tools
• Thinking Snowflake manages network security
• Assuming Snowflake is for designing websites

2. Which of the following is the correct way to describe Snowflake's architecture?

easy

A. Snowflake uses databases to hold data and warehouses to run queries

B. Snowflake uses virtual machines to host websites

C. Snowflake is a peer-to-peer file sharing system

D. Snowflake is a blockchain platform

Solution

1. Step 1: Identify Snowflake's architecture components

   Snowflake separates storage (databases) and compute (warehouses) for queries.

2. Step 2: Eliminate unrelated options

   Options B, C, and D describe unrelated technologies like web hosting, file sharing, and blockchain.

3. Final Answer:

   Snowflake uses databases to hold data and warehouses to run queries -> Option A

4. Quick Check:

   Architecture = Databases + Warehouses [OK]

Hint: Snowflake splits storage and compute: databases and warehouses [OK]

Common Mistakes:

• Mixing Snowflake with web hosting or blockchain
• Confusing compute with storage roles
• Thinking Snowflake is a file sharing system

3. Consider this Snowflake feature: You can increase or decrease resources anytime and pay only for what you use. What is this feature called?

medium

A. Fixed provisioning

B. Static allocation

C. Manual backup

D. Auto-scaling

Solution

1. Step 1: Understand resource scaling in Snowflake

   Snowflake allows dynamic adjustment of compute resources based on demand.

2. Step 2: Match feature to correct term

   Auto-scaling means resources adjust automatically; fixed provisioning and static allocation do not allow this flexibility.

3. Final Answer:

   Auto-scaling -> Option D

4. Quick Check:

   Dynamic resource adjustment = Auto-scaling [OK]

Hint: Dynamic resource changes = Auto-scaling [OK]

Common Mistakes:

• Confusing auto-scaling with manual backup
• Thinking fixed provisioning allows dynamic scaling
• Mixing static allocation with pay-as-you-go

4. You wrote a query in Snowflake but it runs very slowly. Which of these is a likely cause?

medium

A. The warehouse size is too small for the query workload

B. Snowflake does not support SQL queries

C. Data is stored on your local computer

D. Snowflake requires manual server restarts before queries

Solution

1. Step 1: Analyze query performance factors

   Warehouse size controls compute power; too small means slower queries.

2. Step 2: Check incorrect statements

   Snowflake supports SQL, stores data in cloud, and does not need manual restarts.

3. Final Answer:

   The warehouse size is too small for the query workload -> Option A

4. Quick Check:

   Small warehouse = slow queries [OK]

Hint: Small warehouse slows queries; increase size [OK]

Common Mistakes:

• Believing Snowflake lacks SQL support
• Thinking data is stored locally
• Assuming manual restarts are needed

5. You want to analyze a large dataset quickly but only pay for the time you use. How should you configure Snowflake to achieve this?

hard

A. Store data locally and run queries on your computer

B. Use a large warehouse and pause it when not running queries

C. Use a small warehouse and keep it running 24/7

D. Manually copy data to multiple warehouses

Solution

1. Step 1: Choose warehouse size for fast analysis

   A large warehouse provides more compute power for quick queries.

2. Step 2: Manage cost by pausing warehouse

   Pausing warehouse when idle stops billing, so you pay only for usage time.

3. Final Answer:

   Use a large warehouse and pause it when not running queries -> Option B

4. Quick Check:

   Large + pause = fast and cost-efficient [OK]

Hint: Big warehouse + pause = speed and pay only when used [OK]

Common Mistakes:

• Keeping small warehouse always running wastes time
• Storing data locally defeats cloud benefits
• Copying data manually is inefficient and costly