Snowflake Review

Exceptionally good technology that addresses data warehousing challenges and is built and designed in a good way


What is our primary use case?

It is used in my company as well as in my client's company. We are a system integrator, so naturally, we need to have the centers of excellence and competencies in Snowflake.

What is most valuable?

The way it is built and designed is valuable. The way the shared model is built and the way it exploits the power of the cloud is very good. Certain features related to administration and management, akin to Oracle Flashback and all that, are very important for modern-day administration and management.

It is also good in terms of managing and improving performance, indexing, and partitioning. It is sort of completely automated. Everything is essentially under the hood, and the engine takes care of it all. As a data warehouse on the cloud, Snowflake stands strong on its ground even though each of the cloud providers has its own data warehouse, such as Redshift for AWS or Synapse for Azure.

What needs improvement?

There are three things that came to my notice. I am not very sure whether they have already done it. The first one is very specific to the virtual data warehouse. Snowflake might want to offer industry-specific models for the data warehouse. Snowflake is a very strong product with credit. For a typical retail industry, such as the pharma industry, if it can get into the functional space as well, it will be a big shot in their arm.

The second thing is related to the migration from other data warehouses to Snowflake. They can make the migration a little bit more seamless and easy. It should be compatible, well-structured, and well-governed. Many enterprises have huge impetus and urgency to move to Snowflake from their existing data warehouse, so, naturally, this is an area that is critical.

The third thing is related to the capability of dealing with relational and dimensional structures. It is not that friendly with relational structures. Snowflake is more friendly with the dimensional structure or the data masks, which is characteristic of a Kimball model. It is very difficult to be savvy and friendly with both structures because these structures are different and address different kinds of needs. One is manipulation-heavy, and the other one is read-heavy or analysis-heavy. One is for heavy or frequent changes and amendments, and the other one is for frequent reads. One is flat, and the other one is distributed. There are fundamental differences between these two structures. If I were to consider Snowflake as a silver bullet, it should be equally savvy on both ends, which I don't think is the case. Maybe the product has grown and scaled up from where it was.

For how long have I used the solution?

I have been using this solution for close to three years. I kept a tab on Snowflake and its progress since it came into the market.

Which solution did I use previously and why did I switch?

Personally, I have worked extensively with Oracle, SQL Server, and Teradata. SQL Server has the Fast Track Data Warehouse (FTDW) appliance. Oracle has both the database and the appliance. I haven't worked on Parallel Data Warehouse, which is a big one offered by Oracle. Teradata is an appliance in itself. There is also Metadata. I haven't worked on DB2. 

All of these had their own lacunae. Data warehouses had their own problems. There were failures, challenges, and difficulties in adoption, and all of these have been addressed by Snowflake a big way. It has tried to marry the best of both worlds in terms of turnaround time, scalability, adoption, and seamlessness.

I hail from a classical data warehouse background. Snowflake has been kind of a silver bullet. It is trying to meet the best of both worlds. I wish I could do much more on Snowflake, but I'm tied up with many other things, which is why I'm not able to concentrate that much, but it is an exceptionally good technology.

How was the initial setup?

Its initial setup is very simple, which is its plus point. It is not at all a problem. You only need to understand a bit of the cloud ecosystem. When Snowflake is on Azure or AWS, you need to understand

  • What exactly is happening?
  • How these two are handshaking with each other?
  • What part Snowflake is playing?
  • How Azure or AWS is complementing it?

If these things are clear, the rest shouldn't be a problem.

What other advice do I have?

This could be something that might be debated upon, but Snowflake has two parts to it. One is the data warehouse itself, and the other one is the cloud. It is important to know about the cloud in terms of:

  • How a cloud functions?
  • How a cloud orchestrates through its services, domains, invocation of services, and other things?
  • How a cloud is laid out?

For example, let's take AWS. If AWS is invoking Lambda or something else, how will S3 come into the picture? Is there a role of DynamoDB? If you're using DynamoDB, how would you use it in the Snowflake landscape? So, cloud nuances are involved when we speak of Snowflake, and there is no doubt about that, but a more important area on which Snowflake consultants need to focus on is the core data warehousing and BI principles. This is where I feel the genesis of Snowflake has happened. It is the data warehouse on the cloud, and it addresses the challenges that on-prem databases had in the past, such as scalability, turnaround times, reusability, adoption, and cost, but the genesis, principles, and tenets of data warehousing are still sacrosanct and hold good. Therefore, you need the knowledge or background of what a data warehouse is expected to be, be it any school of thought such as Inmon school, a Kimball school, or a mix. You should know:

  • Data warehouse as a discipline.
  • The reason why it was born.
  • The expectations out of it in the past.
  • The current expectations.
  • What being on the cloud would solve?

These things on the data warehouse side need to be crystal clear. The cloud part is important, but it is of lesser essence than the data warehouse part. That's what I see, personally, and I guess that's the way the Snowflake founders have built the product.

As a data warehouse, I would rate Snowflake an eight out of ten.

**Disclosure: My company has a business relationship with this vendor other than being a customer: reseller
More Snowflake reviews from users
Learn what your peers think about Snowflake. Get advice and tips from experienced pros sharing their opinions. Updated: July 2021.
523,431 professionals have used our research since 2012.
Add a Comment
ITCS user
Guest