Migrate to the Cloud

THE HOW AND WHY OF MODERNIZING YOUR DATA WAREHOUSE

What’s inside:
3 What the market says

4 Part 1–Migration framework—approaches, strategies

and requirements

6 Strategy requirements

8 Three approaches for implementing your migration

10 Part 2–Planning your migration

14 Part 3–Taking advantage of your new, cloud

data warehouse

16 Conclusion

16 Learn more

17 About Snowflake
 What the market says
According to Gartner, 92 percent of Enterprises realize that legacy data
organizations are aware they have unmet warehouses can no longer deliver on their
data management demands in support true purpose to organize data, enable
of their analytics.1 Seventy-five percent rapid analysis and make insights available
of data executives say they can’t deliver to all business users who need them.
useful data analytics effectively to the That’s why they’re moving away from
enterprise due to inflexible computing traditional data warehouse solutions
solutions². As a result, more than one- toward cloud solutions.
third of data professionals say their
With some upfront planning and
organizations are already using a cloud
consideration, migrating your data analytics
data warehouse3. But this doesn’t mean
to the cloud is a process that can lead
they want a “cloud-washed” version of
to big payoffs for your business and
a legacy and inflexible on-premises data
technology demands. In this eBook, we’ll
warehouse. Nearly all data professionals,
address your organization’s data analytics
93 percent, see the unique benefits of a
needs with a roadmap for migrating your
data warehouse built from the ground up
data warehouse to the cloud.
for the cloud4.

1 Survey Analysis: New Data and New Analytics Are All Mythology Unless You Add Skills, Gartner.com, 9/18/17
2-4 Data Analytics: Beyond the Hype. A Survey of Data Professionals and Executives, Dimensional Research, 9/16

MIGRATE TO THE CLOUD 3

 1
Migration framework—
PA RT

approaches, strategies
and requirements
There are many reasons organizations THE FOUR MOST COMMON
choose to embrace cloud computing. But MIGRATION SCENARIOS
most organizations need a plan, something The type of migration you embark on
to grab onto and see what the future will significantly influence your migration
looks like. Not everyone is in the same strategy. Here are four potential paths many
place regarding their analytic capability organizations take to migrate their data
or cloud maturity. Therefore, give careful analytics and data warehouse to the cloud:
consideration as to how fast and how
much legacy code you want to move from 1. OLTP for operational
your on-premise environment to a public reporting and analytics
cloud infrastructure. This is extremely common. Many
organizations use OLTP (online transaction
processing) systems, such as SQL Server,
Oracle, or MySQL for basic reporting
and analytics. While this might work
as a short-term solution, the reporting
needs of the business compete with the
operational needs, overtaxing a fixed
resource and slowing performance for
both. A truly elastic cloud data warehouse
eliminates this problem. It’s pretty easy.
As discussed later, take your existing
transactional schema, which is usually in
3rd normal form, and move it, as is, to the
cloud. This removes the reporting workload
from the existing system and houses the
data in a platform built for analytics. This
eliminates the performance bottlenecks,
and in some cases, gives your operational
data store new life.

MIGRATE TO THE CLOUD 4

2. Appliance-based 3. Data marts but no
data warehouse data warehouse
Many of the on-premise appliance vendors Most organizations suffer because
are sunsetting these legacy systems. a single source of truth is always out of
More importantly, their customers want reach. They have data siloed in many
to escape the performance, cost and repositories. They may have tried to
other limitations of these systems that federate access across these repositories
can’t address the ever-changing data but quickly realized the cost to create
analytics needs of the modern enterprise. and maintain that access wasn’t feasible.
Appliance-based data warehouses also They need a centralized repository to 4. Data lakes—data in,
require huge upfront costs typically in the eliminate these barriers to getting all the no insight out
form of capital expenditures (CAPEX). insight from all their data. A cloud-built
Data lake initiatives have become a
data warehouse becomes an obvious
On the flipside, a data warehouse built proverbial black hole: easy to get data in,
choice for data consolidation because
for the cloud is designed to grow with complex but impossible to get data out.
it’s ACID-compliant (transactionally
you from zero upfront costs thanks to Many enterprises have realized that
consistent), can be partitioned/segmented
a pay-as-you-go model, representing an on-premise Hadoop infrastructures are
logically without replicating, and can scale
operational expenditure (OPEX). This costly and complex, and don’t meet their
computing resources up and down, and
removes the guesswork of planning for analytics and concurrency requirements.
on-demand. Whether you’re a Kimball
your biggest day of consumption and Fortunately, there is a path forward.
or Inmon fan, you need a platform that
then overpaying for an underutilized Leveraging a “layered” or “zones” approach
separates compute from storage and
system for the other 364 days of the is a great way for an enterprise to identify
allows end users the greatest flexibly to
year. Similarly, if you need to expand your data sets they can comfortably move to
access data sets using enterprise tools
analytics unexpectedly during the year, the cloud. This method makes it easy to
that leverage ANSI SQL.
you are hamstrung by a system that can’t show the movement of data from on-premise
dynamically adjust to meet your needs. to the cloud in a controlled and secure
Another benefit is near, real-time access manner into cloud storage infrastructures
to data. Since on-premise appliances such as Azure Blob Storage and AWS S3.
are a fixed resource, data warehousing
teams create overnight load windows to
make data available for the next morning.
Today’s cloud-built technologies allow
you to segment workloads and load data
24/7 without impacting query processing,
speeding the time-to-value of your data.

MIGRATE TO THE CLOUD 5

Strategy requirements
Migrations aren’t much different than objectives. Determine if the project is
most IT projects, which means they the “tip of the spear”, a way to get the
usually begin with requirements. Defining company moving towards the cloud in
requirements often cross multiple the right direction. Or, part of a larger
boundaries since a cloud migration cloud initiative which would allow sharing
strategy can be an executive-level best practices and technical resources.
decision. Without executive buy-in, your One of the critical success factors will be
project will be limited in scope and be determining which data sets are ready to
labeled as “shadow IT.” This might work leave your data center first and which data
for some lines of business to get started. sets will follow later to the cloud.
Eventually, everyone needs to be on the
same page, from the architecture team to TECHNICAL STRATEGY
the security team to the chief data officer Every strategy starts with the same
and even the CFO. question:. “What are we trying to do?” For
technology people, this can be defined in
ORGANIZATIONAL STRATEGY the requirements. There is an old adage
For many organizations, cloud is in their that says, “You can have it fast, good or
DNA and even written into their mission cheap. Pick two.” Using a combination of
statements. For others that have been agile development, and taking advantage
around awhile, they know they need to of what the cloud offers, that adage is
modernize but not at the expense of more antiquated than accurate.
changing too quickly. The business benefits
of the cloud are hard to ignore — more
agility, lower costs, deeper analytics. But
your cloud migration project should move
at a pace consistent with your corporate

MIGRATE TO THE CLOUD 6

BUSINESS/FUNCTIONAL NON-FUNCTIONAL
REQUIREMENTS REQUIREMENTS
Design with the end in mind. Discuss Often called the “ilities,” take stock of
goals with existing analytics users and policies related to service level agreements
understand their current challenges and (SLAs) between IT and the business, the
their wish lists. The cloud allows for new security requirements for protecting
capabilities such as near real-time data your data, usability requirements of your
access, data democratization and next-level end users and many other requirements.
analytics with access to detailed data, Common topics include:
not just aggregates. Create a plan or a
i. Security
vision statement that highlights being an
enabler for all lines of business with the ii. Reliability
appropriate security controls and tools. iii. Performance
Use these goals to align IT and lines of
iv. Maintainability
business (LOB) to help set the vision to
securely get accurate information to the v. Scalability
right people at the right time.
vi. Usability

With the cloud, don’t be afraid to include

some aspirational requirements. Once you
have a solid list, label them as either nice-
to-have or must-have. Be wary of paralysis
by analysis, and choose an implementation
window and development methodology —
agile, waterfall, etc. — that is comfortable for
your organization. Pay particular attention
to your high availability and disaster recovery
(HA/DR) requirements. A cloud-built data
warehouse can save your organization from
having to design elegant but expensive
solutions to meet the needs of the business.

MIGRATE TO THE CLOUD 7

Three approaches for
implementing your migration
Now it’s time to figure out what type of Number five is debatable because
migration would make sense. You have many cloud migrations require technical
options, which include lift and shift, lift, changes and changes to the culture of an
“improve” and shift, and full redesign. The organization. If your first initiative doesn’t
steps you’ve taken prior to this stage, provide more that just the same features
such as aligning the migration with your and functions as your old system, can your
organization, and the technical and organization view it as a “win”? Sometimes,
business strategy, will drive your chosen just showing you can migrate without risk
migration strategy. to the business, while improving
performance in some way, can be good
LIFT AND SHIFT enough for your stakeholders.
Most would consider this to be the safest
and most straightforward way to do a LIFT, “IMPROVE” AND SHIFT
migration. The plan is simple: Everything This is by far the most popular approach
we do with the existing system should be and can help bridge the gap between
exactly the same in the new system with the old world of data warehousing and
minimal changes. A lift-and-shift strategy is the new era of big data. The concept is
a good one if: simple. As you’re converting assets, look
for opportunities to streamline or improve
1. Requirements are narrowly defined
the data pipeline, how data is organized,
(very few new requirements).
when data is transformed and how data
2. Time-to-implementation is critical (you is accessed. Then, find ways to take
need to get off the old system ASAP). advantage of new capabilities/functions
3. Your new system has all of the features in the system to which you’re migrating.
and functions of the old system. The theory isn’t to change any of the core
functionality of the system but simply take
4. Your ecosystem of surrounding tools
advantage of the opportunity to simplify
(ETL, BI, system management) requires
or streamline.
minimal or no changes.
The benefit here is to show some
5. The migration is not the centerpiece
improvements over the existing process
of your cloud migration strategy
without breaking things and introducing
(see below).
too much risk. The executive team can

MIGRATE TO THE CLOUD 8

use the migration as a proof point to the RE-DESIGN/RE-ARCHITECT/ warehouse even creates a new opportunity
business. Even though this represents CONSOLIDATE for modern data sharing across and
a major shift in IT philosophy, it does so Many organizations do not have an outside an enterprise. Projects such as
without negatively impacting performance enterprise data warehouse or data lake. this are a great way to consolidate
and provides additional business benefits, In some cases, they’ve been disappointed infrastructure and get a better handle
such as: by their attempt to create one. Their on contracts, security and shadow IT,
data sits in multiple, on-premise systems: while producing incremental results for
1. Faster access to more data
some used for OLTP, some used for the business.
2. More granular data analytics
OLAP and some data sits in file systems Most consolidation efforts start by
3. Better performance on individual just waiting to be analyzed. Changing combining multiple data sets onto the
queries/reports platforms is viewed as an ideal time to new cloud platform to show their analytic
re-architect, or architect for the first time value. The next step would be to gradually
4. No contention for near-unlimited
a fully functional data platform capable of restrict access to the legacy systems,
computing resources
scaling with the business. while growing the capability of the cloud
The platform must meet the requirements data warehouse and establishing quick
outlined in the above sections: handles wins. Many of these initiatives also
multiple types of data, and allows end focus on creating new revenue streams,
users to use their favorite tools and shrinking or eliminating data pipelines and
language (SQL) to create a data consolidating disparate data repositories.
democratization strategy for all lines of
business. In some cases, the cloud data

MIGRATE TO THE CLOUD 9

 2
Planning your migration
PA RT

Executing these steps, and in this order, STEP 2: DOCUMENT THE “AS IS”
is not a necessity. Depending on the This isn’t the most glamorous part of a
scope of your migration, you may need migration but it’s likely one of the most
more or fewer steps. The key is to design critical. You’ll need to communicate both
a framework and core elements of the internally and externally, and up and
plan that you can work from. Assess down the reporting chains, regarding the
your internal skill sets, don’t be afraid to current “as is” implementation. A short list
leverage the best practices outlined by of assets to migrate include but are not
strategic vendors, and consider partnering limited to:
with migration experts.
1. All sources that populate the
STEP 1: DETERMINE THE SCOPE existing systems

Stating the obvious, no two migrations are 2. All database objects (tables, views,
the same and rarely is the end state well users, etc)
understood. The goal is to create a plan
3. All transformations, with schedules
that aligns with the goals of the business,
for execution, or triggering criteria
provides capabilities in the shortest
reasonable timeframe and sets you on 4. A diagram from the interaction of
the path for incremental improvement. systems/tools
Your end state could be getting a single
STEP 3: DETERMINE THE
workload into the cloud within one
APPROACH AND ASSEMBLE
month, or it could be migrating your entire
THE IMPLEMENTATION TEAM
analytics platform by the end of the year.
It’s reasonable to plan for a one-year ROI, Multiple options exist here. We've outlined
which you can even accelerate under above the most common approaches
certain scenarios. but there are combinations of these. You
could choose to implement one method
to get to initial capability and another as
you approach full production. Creating
high-level milestones at this step is a good
way to segment when a capability will be
available, and which requirements you’ll
satisfy via release schedules.

MIGRATE TO THE CLOUD 10

STEP 4: DIAGRAM THE “TO BE” business and executive audiences, and
Once you get your arms around what each will want different levels of detail
you want to migrate and when you want on the initial operating capability (IOC)
capabilities available, you can begin to and final operating capability. Below is
document the “to be” architecture. Be a “to be” example of modern cloud data
aware, there is no “one chart to rule them warehouse architecture:
all”. You’ll be communicating to technical,

NO ETL
ETL A MOD ERN CLOUD
ELT
D ATA WAREHOUSE
DATA FLOW
ARCHITECTURE

S3 /
AZURE
STAGING NATIVE
DATA BASES CONNECTOR /
ODBC / JDBC

VIRTUAL ANALYTICS &

WAREHOUSE DATA SCIENCE
FOR DATA USERS & TOOLS
REPLICATION SCIENTISTS
& STREAMING
TOOLS VIRTUAL
STAGING TABLES
WAREHOUSE
FOR LOADING
WEB UI

VIRTUAL AD-HOC SQL

WAREHOUSE QUERIES
SOURCE
FOR AD-HOC
SYSTEMS
USERS
(CLOUD OR
ON-PREM)
VIRTUAL
NATIVE
WAREHOUSE FOR REPORTING CONNECTOR /
TRANSFORMATIONS
DATABASE(S) ODBC / JDBC
DATA (SCHEMAS, TABLES,
TRANSFORMATION VIEWS, ETC)
VIRTUAL BUSINESS
TOOLS
WAREHOUSE(S) INTELLIGENCE
FOR BI TOOLS TOOLS

MIGRATE TO THE CLOUD 11

STEP 5: PLAN FOR YOUR DATA data warehouse. Some sources and
LOAD AND “SIZING” ETL tools support change data capture
One of the most challenging aspects of (CDC) strategies. Others might support
migrating to the cloud is moving data and all inserts, while others might require a
changing your paradigm to take advantage full refresh. There is no one size fits all
of the elastic resources of the cloud. There approach. If you have the requirement that
are three pieces to the puzzle: all data must end up in the lake, you can
either write to the lake first or the data
1. Initial load warehouse first. It’s your choice. This is a

The initial load can be challenging place where you could limit the amount of STEP 6: CONVERT ASSETS

based on data volumes and security change in the existing process and revisit This step refers to defining data
requirements. Work closely with your after the initial implementation. warehouse/database assets you may
security team, and the lines of business need to convert. These include data
3. Planning for warehouse
that own the data to make sure you definition language (DDL), role-based
usage and storage
don’t have to go through a tokenization/ access control (RBAC) and data
obfuscation process before moving Typically, most organizations execute manipulation language (DML) used in
data into the cloud. Many organizations a POC and go through an ROI exercise scripts. The good news is that most
segment their data into “zones” or “layers” before executing a migration. At this phase, relational databases leverage the
inside their data lake: raw, curated, it’s usually a good idea to re-validate the ANSI-SQL standard. Most of the changes
aggregated and cleansed areas. Then make usage plan and work the operational side will revolve around ensuring DATE and
the decision which data sets are okay to of the equation with regards to how to TIMESTAMP formats are converted
move, taking into account regulatory and monitor availability and how to govern correctly, and the SQL functions used to
data privacy compliance standards such as usage of the system. Because some cloud access those are checked for compliance.
PII, PCI and HIPAA. Regarding the volume data warehouses provide the ability to (Not all vendors implement functions
of data, as networking gets better, this scale up and down, turn resources on and the same). Some cloud data warehouses
issue starts to go away as you can move off, segment workloads and resources, and simplify DDL by eliminating the need to
terabytes of data into the cloud. auto-scale both processing power and partition and index, so your DDL becomes
storage, the model changes from time much cleaner (less verbose).
TIP: Many organizations receive data
slicing a fixed resource (limiting your
externally from partners and vendors STEP 7: SETUP YOUR “TO BE”
business user access) to allocating resources
(Salesforce, etc). It might be prudent to ENVIRONMENT AND TEST
based on business need and value. You no
dump these data sets into a cloud object CONNECTIVITY / SECURITY
longer have to do a big planning exercise
storage service to keep it from becoming
to handle your largest workload and leave It should be no surprise that you’ll have
classified as an on-premise asset. If data is
the system underutilized for the other 364 to complete your networking, proxy
coming over the Internet, it should be ok
days of the year. and firewall configurations during your
to secure it in the cloud.
TIP: If you are integrating your migration migration strategy. It usually helps to have
2. Ongoing updates with your data lake strategy, be aware of a chart or two outlining what ports and
transfer charges of moving data between URLs you will need to access. You will also
Each source of data, the ETL logic and
regions or cloud providers. want to work with your security group to
integration with the data lake strategy,
download and install any drivers (ODBC,
will dictate the methods used for
JDBC, etc) or support software such as
updating data in your cloud enterprise
a command line interface (CLI), which

MIGRATE TO THE CLOUD 12

most DBA-type developers prefer to use STEP 10: RUN YOUR EXISTING AND
when interacting with a modern cloud NEW SYSTEMS IN PARALLEL
data warehouse. You will also want to set When replacing an enterprise data
up your account parameters such as IP warehouse or data lake, it’s easy to dual
whitelisting and role-based access control load the systems and run them in parallel.
before opening the environment up to From there, gradually move users or
larger groups. groups onto the new system, trying not
to disrupt business operations. Target
STEP 8: TEST THE PROCESS END TO
groups that expressed the most challenges
END WITH A SUBSET OF DATA
or concerns during the planning and
Once you have connectivity worked out requirements phase as they will be the
for all tool sets, it’s best practice to test most receptive and probably become the
your process from end to end for both most vocal advocates about success.
functionality and performance. If you’re
coming from an existing system, you’ll STEP 11: PICK YOUR
have the advantage knowing what your CUTOVER DATE
service level agreements (SLAs) are with You can run the systems in parallel for a
each line of business and requirements for while but once your end users experience
each step in the process – load, transform/ the new performance benefits they will
aggregate, query execution, etc. It’s always never want to go back. You’ll likely want to
best practice to select test cases and run the systems in parallel for at least one
data sets critical to early success. This major reporting cycle—a week, a month or
is also an opportunity to implement the a quarter. Once you pick an official cutover
“improve” part of the migration, if that’s date, continue to dual load the systems
the methodology you chose. in case you run into a problem. Once
everyone seems happy, it’s time to pop
STEP 9: MIGRATE THE DATA AND
the champagne and retire the old system.
TEST PERFORMANCE
Before going live, you’ll want to re-run some
or all of your performance tests from step
eight to ensure the system is configured
for individual query performance (size of
the warehouse) and for concurrency
(scaling out the warehouse). You’ll want a
champion from each of the critical lines
of business to test accessing the system,
making sure their tools work and ensuring
they get the performance you expect for
them. Validate that you’re getting the
same calculated results from the old to the
new system.

MIGRATE TO THE CLOUD 13

 3
Taking advantage of your
PA RT

new, cloud data warehouse

Now that your migration is complete, the ACCESS TO CROSS-BUSINESS
work isn’t done. It’s time to start taking UNIT DATA
advantage of the new capabilities. One Along with the LOB specific data sets,
strategy is to look for ways to improve most organization also want access to the
performance and get data into the hands latest and greatest information from other
of end users more quickly, or distributed business units and sources. Some modern
to more users. Some options include: cloud data warehouses provide several
methods for controlling data access while
DETERMINE LOB AND
still enabling curated data sets to other
END-USER NEEDS
end users. In some cases, organizations
You compiled an initial list of requirements share or receive data sets via FTP with
from the LOBs earlier in the process. Now other organizations. Moving data this way
it’s time to revisit that list and start a dialog is time consuming and expensive, especially
with the LOBs, educating them on what’s as data volumes and frequency increases.
possible with the new system. Some of Some modern data sharing features
these new capabilities include but are not associated with modern cloud data
limited to: warehouses eliminate the need to transfer
a. More data—access to detailed records and transform data, streamlining the process
and reaching back years, not just and reducing ETL cost and complexity
months or weeks
OPTIMIZE (RETHINK) YOUR
b. Different data types—Structured and LOAD STRATEGY
semi-structured (JSON, AVRO, XML,
Typically, enterprise data warehouses
PARQUET, ORC)
have load windows to execute in batch
c. Cleaner data overnight, making yesterday’s data
available for analysis the next morning.
d. Better formatting/modeling—changing
With today’s technology, you can load
schemas from third normal form (3NF)
data and query data without contention,
to star or other data models
opening the possibility to load data 24/7,
e. Faster performance—many and providing data access sooner to end
organizations use summary tables users. You can spin resources up and down
or materialized views to improve instantly, which allows you to load data
performance as fast as you want, and at the same price
point no matter the compute resources
you spin up.

MIGRATE TO THE CLOUD 14

For example, let’s say it takes four hours By separating compute resources from
to load 1TB of data every night. In some storage resources, you have the ability
cases, you’re using one node for this to load data without impacting query
workload. With near linear scalability, if performance, making near, real-time
you execute the same workload with a data processing at scale a reality. You just
two-node cluster, it would load twice as need to figure out how much change your
fast but cost exactly the same. Double the organization can absorb at one time.
cluster size again and it gets done twice as
fast again at exactly the same cost. IDEMPOTENT LOADING
Never heard of it before? This isn’t a
ANALYZE SOURCES FOR correctly), fix step 45 and continue
new term, but has been coming up in
SIMPLIFICATION/STREAMLINING processing. Most legacy platforms
lots of conversations related to big data
would force you to reload or recover
In some cases, the tools/systems you were processing. The goal is simple. You want
the original table and start over. If you
using in the past may not be necessary to continually load data into your system,
need a snapshot of your database, or
in the future. Today’s modern cloud data but if something goes wrong along the
table before doing an update, use
warehouses can handle new data types way you don’t want to get confused about
a zero-copy clone function to make a
and process data more efficiently, blurring what has been processed and what hasn’t.
copy of either but without duplicating
the lines between OLTP, OLAP and the Idempotent loading means that it doesn’t
the data. That’s a big deal, especially if
data lake. Large organizations tend to matter if you load a file or record once or
you’re currently doing large ETL jobs to
have many tools, all bought for specific ten times, you end up with the exact same
pull terabytes of data into data marts for
functions or capabilities. result. This is pretty powerful and can
data science or test/QA teams to work
be achieved using a combination of the
Now is good time for an ETL/ELT tool with. And finally, did you accidentally
COPY command and transformation using
rationalization strategy. Some of these drop a production-level database or
the MERGE commands found in modern
vendors have upgraded their products to table while doing the nightly change?
data processing platforms. It’s especially
work in a cloud environment. Do not try No worries. UNDROP works really well
good for streaming data with updates or
to standardize on a single tool. Instead, in that situation, and magically all of the
deletes but you can implement it in other
make technical recommendations for data and schema reappears.
situations as well.
accessing specific sources, or, if you are
implementing a data lake strategy, use
ZERO-COPY CLONE, TIME TRAVEL
different tools for accessing different
AND UNDROP
zones or layers.
Some modern cloud data warehouses
You should also analyze the tools in the have the ability to query “back in time”,
context of how quickly they can move data between the previous 24 hours and
from your original system to your new 90 days. Many organizations use this
platform. Most organizations are moving capability to transform the ELT pipelines.
from extract, transform and load (ETL) So, if something goes wrong in step 45 of
strategies to extract, load and transform a 50-step process, you don’t have to start
(ELT) and stream processing strategies to over. You use time travel to set yourself up
take advantage of on-demand scaling. to a known state (step 44 that executed

MIGRATE TO THE CLOUD 15

Conclusion Learn more
MORE AND MORE DATA IS Click here to get more information about
BORN IN THE CLOUD how to modernize your data warehouse

Enterprises now realize there are efficiency and to get instructions specific to migrating

and performance advantages to storing, from your existing, on-premises data

analyzing and sharing data in the cloud. warehouse to Snowflake.

This approach removes numerous steps

for enterprises, and removes the chaos
that can result from siloed, duplicated data
that becomes disconnected from its original
source. A carefully planned migration
can lead to significant advantage over
conventional data warehouses, including
more capabilities at lower cost.

MIGRATE TO THE CLOUD 16

About Snowflake
Snowflake is the only data warehouse built for the cloud, enabling the data-driven enterprise with
instant elasticity, secure data sharing and per-second pricing, across multiple clouds. Snowflake
combines the power of data warehousing, the flexibility of big data platforms and the elasticity of
the cloud at a fraction of the cost of traditional solutions. Snowflake: Your data, no limits. Find out
more at snowflake.net.

MIGRATE TO THE CLOUD 17