Frank Bell, Data Thought Leader and Snowflake SME at Accenture - CEO at ITS We will cover all aspects of optimizing your Snowflake Data Cloud including: *Dive deep into how Snowflake pay as you go costs work and how by utilizing our proven optimization tools - Snoptimizer SaaS Snowflake Optimizer - https://snoptimizer.com/ , scripts, and architecture techniques you typically can save 10-40++% on your existing Snowflake Account costs. *Explain how Snowflake Compute works and proven techniques on how to architect warehouses for both cost and performance efficiency. We cover in depth how snowflake scales BOTH out and in as well as up and down with compute resources. *Explain how Snowflake data storage works with Replication, Time-Travel, and Cloning. We explain these awesome features as well as their downsides if they are used and configured wrongly. *Cover Snowflake cloud services costs and features that have costs related to them, including Snowpipe, Search Optimization, Materialized Views, Auto-clustering, and other recent new cost based features that provide value at a cost. *Finally, we will discuss how you can ensure your Snowflake Account(s) are fully optimized not just for cost but also for security and performance on Snowflake. We will show you security and performance best practices as well as pitfalls to avoid.

1. Snowflake Data Cloud
Optimization Fun!
How to Optimize your
Snowflake Data Cloud
Best Practices and Tips
from a Snowflake Data
Superhero

2. • Intro – Speaker and the Snowflake Data Cloud
• Snowflake Evolution and Difference
• Snowflake Consumption Pricing
• Snowflake Optimization
– Cost
– Performance
– Security
• Optimization Techniques
– Hard Way
– Easy Way (Automation Tools)
• New Cost and Resource Monitoring Tools coming in 2022
– Resource Groups
– Budget Assignment
DATACONLA – 2022 - Agenda

3. Frank Bell
*Main author of the book Snowflake Essentials
*Top Snowflake Data Thought Leader at Accenture
*Started Snowflake LA Users Group 2018
*Ran top Snowflake Data Cloud Consulting Practice
[acquired by Fairway/Accenture in 2019]
*25+ years of data .. “stuff” 
*Created Snowflake Solutions and Snoptimizer
https://snowflakesolutions.net – Snowflake Business User and Developer Community with
Knowledge Repository and Snowflake Tools
https://snoptimizer.com – Automated Snowflake Data Cloud Optimization Service
Snowflake Data Superhero
*Creator of Snowflake Solutions and Snoptimizer

4. • Snowflake Data Cloud is:
• A cloud based database and
interconnected data system
that can handle multiple
workloads
• Fully connected data cloud
with “no-copy” data sharing,
data cloning, and data usage
enabled within a cloud
provider region.
What is the Snowflake Data Cloud?
KEY POINTS:
*Unified system
*Connects companies and data
providers to data
*Single and seamless experience
*Run across multiple public cloud
*Removes massive amounts of
friction from data access and
processing

5. • 2014-2018 – Snowflake Database (Focused on Data
Warehouse)
• Structured and Semi-Structured Data
• 2019-2020
• 2021 Snowflake Data Cloud
• Workloads: Data Warehousing, Data Lakes, Data
Applications, Data Science, etc.
• Data Types: Structured,Semi-Structured,Unstructured
Snowflake Evolution

6. Frank’s Snowflake Differentiators: (from all other data systems)
#1: Architected on cloud providers separated storage
and compute.
#2: Micro-Partition Architecture
#3: ”no-copy” data cloning – enables DataOps and
true Agile Data Systems/Applications
#4: “no-copy” data sharing – “game changer”
Why is the Snowflake Data Cloud
so different? (Part 1)

7. Frank’s Snowflake Differentiators: (continued)
#5: Capability to process Structured, Semi-Structured
and fully Unstructured data all in one system
#6: End to End data processing and data science
workloads in one fully secure and governed system.
#7: in progress
#8: in progress
Why is the Snowflake Data Cloud
so different? (Part 2)

8. • Overall Snowflake is Awesome
but it is not continuously optimized
for cost, security, or performance
• Snowflake continues to add excellent features and many
of them improve performance but also adds cost,
security, and performance complexity. -
Snowflake
Consumption Based Pricing

9. Consumption Based
Pricing is Awesome!
Unless you have:
1. no optimizations
2. resource constraints!
Some Snowflake
Customers who have a
budget of $50,000/year
can blow through it in 3
days without
optimization
Medium to large size
customers who do not
optimize extensively can
miss out on saving
$5,000+/month
Automated Snowflake Data
Cloud Optimization is what
keeps Consumption Pricing
on Snowflake Awesome!
Then it's not …

10. Resource Monitors, warehouse
Optimization, etc.
Cost
Roles, Warehouses, Privileges, Grants,
Stale Users, Network Policies, and 30
more tests
Security
Queuing, Spilling, etc.
Performance
Snowflake Optimization Areas

11. Resource Monitors, warehouse
Optimization, etc.
Cost
Roles, Warehouses, Privileges, Grants,
Stale Users, Network Policies, and 30
more tests
Security
Queuing, Spilling, etc.
Performance
Snowflake Cost Optimization
Best Practices

12. Most costs found here
Query Consumption
Typically should be low
Automatic Clustering
What out for stages & time
travel
Storage Consumption
Need to monitor
Search Optimization
Need to monitor
Materialized Views
Typically low but you still need
to monitor
Cloud Services
Typically low - very efficient
Pipes
Need to monitor if you have
replication set up
Replication
Cost Areas on Snowflake

13. Before After
Customer A Profile
• Capacity Contract -$100,000/year
• Data Platform Size: 10+ TBs
• Uses Snowpipe for IOT data
• $12,000+/month consumption costs
Optimization Results
• Cost Optimization around Load
Warehouses with non-optimized
suspend and parameters setting
• Exact code to fix above issues
• $10,200/month consumption costs
Cost Optimization

14. Tip #1
Cost Optimization
Credit Consumption
Optimization - Using
Resource Monitors
(and soon Resource
Groups)
*Put actual code here
on Resource Monitor
Best Practices

15. Tip #2
Cost Optimization
Warehouse
Optimizations – DO
NOT USE DEFAULTS!
(Auto-Suspend – give
details)
*Place code here

16. Tip #3
Cost Optimization
Cloud Data Storage
Common Cost Issues
*Anti-Pattern
*Put actual code
here?

17. Tip #4
Cost Optimization
x Data Loading
optimization tips and
tricks
*Add actual code here

18. Snowflake Optimization
Techniques
Snowflake Optimization Technique Services Available PROS CONS
*Reporting on Costs. Manual
Optimizations
*link to cost reporting Quick to implement. Not comprehensive at all. Also, very
reactive versus proactive. Major
problems can occur on cost,
performance, and security very quickly.
*Custom Coded automated optimization
system
* More thorough and typically more
proactive than reporting and finding
error culture.
Extremely expensive. Time-consuming.
Easily outdated if large staff of
performance experts is not continuously
used.
*Relying on Human based Consulting or
Snowflake Professional Service Health
Checks
Snowflake Usually engages reasonably detailed
Snowflake Consulting Experts
Expensive and not repeatable.
Prone to human error. Is outdated often
within days of it being finished.
*Fully Automated SaaS services that
automate optimization
Snoptimizer
Nadilytics SaaS
Security [. ]
Automates the tremendous complexity
of optimizing Snowflake for cost,
performance and security continuously.
Medium Cost
?

19. Resource Monitors, warehouse
Optimization, etc.
Cost
Roles, Warehouses, Privileges, Grants,
Stale Users, Network Policies, and 30
more tests
Security
Queuing, Spilling, etc.
Performance
Snowflake Performance Optimization
Best Practices

20. Tip #1
Performance
Optimization
Snowflake Queuing
Optimization
*Code

21. Tip #2
Performance
Optimization
Snowflake Spilling
Optimization
*
Code

22. Tip #3
Performance
Optimization
Data Loading
Optimization
*Code goes here

23. Customer C Profile
• Capacity Contract -$200,000/year
• Data Platform Size: 50++ TBs
• 3TB ++ size table with non-optimized
cluster keys - Many queries running
over several minutes
• Slower External Table Queries
Optimization Results
• Improved Cluster Keys and queries
above 1 minute all fell under 1 minute
• Materialized Views for certain External
Tables - Massive Query Performance
Improvement
Before After
Performance
Optimization

24. Resource Monitors, warehouse
Optimization, etc.
Cost
Roles, Warehouses, Privileges, Grants,
Stale Users, Network Policies, and 30
more tests
Security
Queuing, Spilling, etc.
Performance
Snowflake Security Optimization
Best Practices

25. Tip #1
Security
Optimization
Security Opt
*Add code

26. Tip #2 Security opt
*Add code
Security
Optimization

27. Tip #3 Security opt
*Add code here
Security
Optimization

28. Tip #4 Security opt
*Add Code
Security
Optimization

29. Customer B Profile
• Capacity Contract: $50,000/year
• Data Platform Size: 1-2TBs
• Significant Cost Risk due to a large
amount of users having CREATE
WAREHOUSE at any size granted.
• Stale objects with too many users
having access
• Stages not properly secured
Optimization Results
• Improved RBAC, cost risk significantly
reduced
• No stale objects
• Stages properly secured
• Continuous Security Analysis and
Protection
Before After
Security Optimization

30. Optimization Results
• In our tests, on average we have seen 10-30% cost savings,
1000s of security issues fixed and 100s of performance
problems solved.
• In some implementations we have seen almost 50% cost
savings.

31. Contact Us
Frank Bell, Snowflake Data Superhero [3rd year]
fbell@itstrategists.com
https://snowflakesolutions.net
https://snoptimizer.com
Snowflake Questions?

32. Join LA
Snowflake
Users Group
and Meetup
https://www.meetup.com/los-angeles-business-snowflake-users-group/