Snowflake Reviews

We are silver or gold partners. The main use case is that we are building a data lake. We are creating a couple of downstream applications as well that will be used by data scientists. So, we will have a single data lake that will be used across the organization by different business domain users. The data is multi-source. We have data from SAP, JDE, and some Excel files.

The speed of data loading and being able to quickly create the environment are most valuable. 

For data, it provides built-in security and compliance with different standards, such as SOC 2, ISO, etc. So, we don't have to do a separate audit for compliance.

There are some gray areas. For example, there is no clarity on where the data sits exactly. That is their proprietary information, and they are not sharing those details. 

Its price should be improved. On the cost-side, it is more expensive than others.

If they could bring in some tools for data integration, it would be really great.

I have been using this solution for almost two and a half years.

It is pretty stable. Its stability is excellent.

It is scalable. As of now, there are 500 users, but slowly, we are planning to roll out to multiple regions. It is currently in Europe, and we will be rolling it out to the APAC and USA regions. By the year-end, there will be more than 1,000 users.

They're perfect. They're excellent. It could be because we are partners.

It is straightforward. It is not that complex.

Our own team deploys it for customers, but the initial configuration is done only by the Snowflake team because that is their area. 

I have worked with multiple clouds, and cost-wise, it is a bit costlier than others, such as Redshift. Its price should be reduced.

I would rate it an eight out of 10.

My company wanted to have all our data in one single place and this what we use Snowflake for. Snowflake also allows us to build connectors to different data sources. The ultimate goal is to provide reporting and analytics to all departments in the company.

This solution could be improved by offering machine learning apps. 

We have used this solution for one year. 

This is a stable solution. 

This is a scalable solution. 

Before Snowflake, we used Azure.

I worked with a Microsoft partner to set up the entire thing. This took three months. 

We have not yet experienced ROI. 

Overall, this is a helpful tool with a friendly interface. 

I would rate this solution an eight out of ten.

The most valuable feature is the clone copy.

In a future release we would like to have a link which would allow us to connect to an external database and create certain views in your own database. This is because it is becoming hard for us to compare the data between multiple sources.

I have used this solution for two years.

This is a stable solution. 

This is a scalable solution. 

The technical support of Snowflake is good. 

Positive

We have seen ROI when using Snowflake based on the insights we get.

I would recommend the SaaS version for their organization. It is not complicated to use. Establishing a private link with current cloud services has been challenging so I would recommend having some kind of a block.

I would rate this solution a nine out of ten. 

My current client is migrating from an on-premise data warehouse to the cloud, and they need to consolidate the data lake. I'm using Snowflake to develop the data lake and build data bots for certain functionalities. I've also used Snowflake to aggregate and clean data for my client's analytics.

I like Snowflake's data exchange capabilities. It can exchange data with downstream systems and other vendor partners as well.

There are some challenges with loading unstructured data and integrating some message queues or brokers. In one project, we had a problem connecting to one of the message queues and we had to take a different route altogether on Microsoft Azure. 

I started using this product about two years ago at my last company. Now, I'm a registered Snowflake partner, so working with Snowflake for one of my clients.

My clients haven't faced many stability challenges. It is a relatively new product that's still evolving. They're using it mainly for POCs, so they haven't faced any problems. 

The most valuable aspect of Snowflake is its scalability. The volume of data, which I have seen across almost in terms of five terabytes and other terabytes of data, I didn't see issues.

Snowflake technical support is good. At my previous company, they had a customer success officer for critical accounts, and they were accessible. When I registered as a partner, I also got a quick response from support. I would rate their support eight out of 10.

Positive

Setting up Snowflake isn't complex. I've not seen any challenges.

They have different pricing structures based on data usage. I think they have three or four tiers you can scale through. Most of the clients see an advantage from this type of licensing. That's the reason they go for it. It's priced competitively with other solutions.

I rate Snowflake eight out of 10. 

We are completely migrating to Snowflake, and we are in transition. It is primarily to combine all our data repositories into a single place. We have SAP BW and SAP HANA, and some of our business units have their own databases. We chose Snowflake to consolidate all of our data into a single place and then build enterprise data. We are then going to provide the data for our businesses in shared databases, on which they would do reporting. They will also have the ability to bring in their own data, which is currently not possible. They will also be able to do advanced analytics, machine learning, and AI in Snowflake, which is not fully possible on our current platforms. It will be used for all the operational reporting, such as sales, supply chain, appraising, and merchandising. We just started to do reporting related to sales and supply chain inventory.

We have its latest version. It is currently deployed on Amazon AWS, but we are moving to Google.

There are so many features that Snowflake offers to address the challenges that people have been facing in the traditional data warehouses for a long time. It allows us to have a single repository for all the data. Currently, we have data repositories all over the place, and we want to bring everyone onto one platform so that it can be utilized across the organization. Currently, we need database administrators and SAP administrators to manage multiple databases and platforms. With Snowflake, we don't need any admin, and there is zero maintenance. All we need is a platform architect who can just manage the Snowflake platform to create databases and security roles, and then you can share the data. By integrating everything into a single Snowflake platform, we have lowered the total cost of ownership quite a bit.

The ability to share the data and the ability to scale up and down easily are the most valuable features. The concept of data sharing and data plumbing made it very easy to provide and share data. The ability to refresh your Dev or QA just by doing a clone is also valuable. It has the dynamic scale up and scale down feature. 

Development and deployment are much easier as compared to other platforms where you have to go through a lot of stuff. With a tool like DBT, you can do modeling and transformation within a single tool and deploy to Snowflake. It provides continuous deployment and continuous integration abilities.

There is a separation of storage and compute, so you only get charged for your usage. You only pay for what you use. When we share the data downstream with business partners, we can specifically create compute for them, and we can charge back the business.

They need to incorporate some basic OLAP capabilities in the backend or at the database level. Currently, it is purely a database. They call it purely a data warehouse for the cloud. Currently, just like any database, we have to calculate all the KPIs in the front-end tools. The same KPIs again need to be calculated in Snowflake. It would be very helpful if they can include some OLAP features. This will bring efficiency because we will be able to create the KPIs within Snowflake itself and then publish them to multiple front-end tools. We won't have to recreate the same in each project. 

There should be the ability to automate raised queries, which is currently not possible. There should also be something for Exception Aggregation and things like that.

I have been using this solution for two years.

It is all cloud. It is really stable. We haven't seen any problems.

We can scale up or down based on our needs. We don't have tons and tons of data, but based on the quality feedback from our vendors, it can handle large volumes and has the competency. With the dynamic scale-up feature, we are confident that it is going to meet all our requirements.

Currently, our number of users is very limited because we have just started the migration. We don't have many users on the platform. All of our focus is on Snowflake because we're moving to Snowflake, and its usage will increase in the future.

I do not directly interact with the support, but I believe our platform architect reached out, and he got a response.

We had SAP BW and SAP HANA as our main data platforms. We are slowly decommissioning SAP BW and SAP HANA and completely migrating to Snowflake. We wanted to have a single repository for all the data. The cost was also a factor.

It is straightforward. To expose the data in the cloud, we had to go through our info security and legal, so that's the part that took time. After that is done, the process for setting up the platform, getting signed up with the initial free credits, and signing up the licensing for the credits was straightforward.

We are working with a system integrator or vendor for this project. Our strategy is to work with an experienced vendor for the first project, and after that, we would be able to drive things forward.

Our experience with them is good. They're building the architecture of Snowflake. They have experience, and we have our own thoughts. We are working together and making sure that the architecture is for the long-term and not just for one project. Whenever we see that their focus is limited to the project, we are asking them questions to make sure that they are making the right decision.

In terms of maintenance, it doesn't require any maintenance, but you do require architects. We have three architects. One architect is responsible for the platform and takes care of creating security rules, grants, and users. We also have an integration architect who is responsible for data acquisition, ETL, and stuff like that. We have a data architect who is responsible for the overall data architecture in terms of what layers we need to establish and how do we model the data and publish that for consumption.

There is a separation of storage and compute, so you only pay for what you use. 

The key part is skill set because Snowflake is all SQL-driven data warehousing. Internally, we have some SAP BW development resources, and they need to learn and move on to understanding SQL-based coding and custom data warehousing concepts.

I would rate Snowflake a nine out of ten.

I use it for data warehousing. I just design databases, put data in there, and get data out.

Although I haven't used it much, it is very fast and the performance is great.

It has some technical quirks that whoever is using it needs to be aware of. It doesn't enforce typical relational database constraints. If you're not aware of that, you can really put some bad data in there.

I've been using it for about three months.

I would give it a ten out of ten. I haven't seen it failover.

I've never had any issues with it. It is being extensively used. We have between a hundred and a thousand users using Snowflake.

I have not had any experience with customer support, but I've heard it's good.

I've heard it's quite expensive.

I would give Snowflake a seven out of ten.

I have been working on Redshift, Snowflake, and AWS RDS Oracle. In the particular case of RDS Oracle, they were migrating from on-prem Solaris equipment to cloud-based RDS.

I would suggest Snowflake for anyone with the need for a reporting/business analytics view of their data that wants only wishes to maintain technical FTE's around processing the data into or out of a data repository but, doesn't want to go to extent of technical management of "AWS clusters" for the data repository.

It was relatively easy to use, and it was easy for people to convert to it. Moved 168 tables and appropriate indices to Snowflake with minimum modification to Current Oracle DDL. The largest degree of change was setting up the corresponding access Hierarchy to duplicate what was in Oracle ( customer had separate permission structures for application vs Admin/support vs direct reporting access to the data).

The aspect of it that was more complicated was stored procedures. It does not support SQL language-based stored procedures. You have to write in JavaScript. If they supported SQL language and stored procedures, it would make migration from on-prem much simpler. In most cases, if an on-prem solution has stored procedures, they're usually written in SQL. They're not written as what most on-prem DBMS would refer to as an external stored procedure, which is what these feel like to most people because they're written in a language outside of SQL.

The other thing that people found difficult to deal with was that they had several Oracle DBAs who were very experienced DBAs, but they were used to on-prem. They were used to having the ability to turn any dial and flip any switch. Moving to Snowflake did cause some issues there because they had to completely readdress the fact that they couldn't touch the engine, and they had to spend more time analyzing performance.

I probably used it about six months ago. I haven't been working with a client who is currently on this platform.

I haven't had to call on them for a problem at that level.

It was a cakewalk. The biggest thing that's hard to do with it is that you have to do an analysis of performance over time to determine the scale because they separate compute and storage.

Scaling the query to a proper size compute is the larger aspect of the problem for most people. That's because you're looking at something completely different. The problem is that you're now trying to figure out what is the largest compute you need to keep performance where you want it without going too large. If you were in an on-prem scenario, you would tweak and twaddle all the dials. You might rewrite the query, but at the end of the day, you're still working inside the same physical acquisition or same physical resources, whereas in Snowflake, you're literally saying that you've got a 10 million row table as part of your query, but what is the necessary compute facility that you need to run queries that are running against that table.

It is hard to say because we're usually engaged in the transition as opposed to the long term. Their storage costs are easily within pennies of what AWS S3 would normally cost. 

Most of the clients I've been working with are in the financial sector, and they're relatively small. I would put them in an SMB connection. The first thing we have to bring up for people is that they're going to build this. They shouldn't store their data in S3. They should pipeline directly into Snowflake and use it on their storage. So, the cost is a big issue because these are small to medium size companies, and that is the biggest thing we had to price point for them.

The biggest conversion problem we've seen so far is when someone had a large number of stored procedures that were SQL-based, as opposed to external stored procedures written in C or whatever the DBMS would support. Converting those stored procedures either to a SQL script or to a stored procedure or function that's based on JavaScript is the biggest challenge that most people we've dealt with are having. That's because they have to relearn the language they're writing their logic in.

I would easily rate it an eight out of 10.

I use this solution for actively building out the cloud data warehouse and data platform for enterprise level customers as well as startups. Generally, our clients are looking for a data warehouse on the cloud to enable them to scale infinitely at a lower cost. I've worked for a finance analytical team building their data lake, the data platform on top of Snowflake, as well as for a telehealth team. It's basically about getting data from multiple sources and building out an entire data platform with data governance. We are customers of Snowflake. 

One small company I worked with had a MySQL RTS based instance and were using AWS RDS with MySQL on top of that. As a result they were unable to scale their database because there were around half a million queries being run per second as well as data querying and data updating. The migration to Snowflake helped the company because there are no limitations in the cloud and no longer restrictions on the queries. Performance for end users improved whether they were internal or external clients. They used to sell the data through APIs so this migration helped to grow their business overall as well as the ML team efficiency and the productivity of users who previously used the data platform. 

The most valuable feature of Snowflake is the query performance. Snowflake is the most efficient way for real-time dashboards or analytical business intelligence reports to be sent to the customer. There are a couple of areas where they have recently improved. One of the key features they introduced is an internal, table-based merch as well as storing of the unstructured data. You can now build a table out of unstructured data, metadata. This hasn't yet been officially announced.

Although the UI has improved lately, they still need to work on their UiPath, the workspace area.

I've been using this solution for two years. 

It's an infinitely scalable system, but if you use terabytes or petabytes of data, then you need to tune the levels. Each day, we get four to five gigs and overall, our data warehouse has 100 gigs plus, it's huge data. 

Our clients previously used the RTS based MySQL and migrated to Snowflake from there. The primary reasons they moved was because of scalability and performance. Other than that, Snowflake reduces costs quite significantly. I also have experience with BigQuery which is particularly used for Google Cloud although these days they have a multicloud enrollment. Snowflake is vendor independent so you don't have to stick everything in Google Cloud. In terms of performance, Snowflake is faster than BigQuery. 

The advantage of Snowflake is that it's easy to deploy and they take care of the setup. Basically, it's a cloud warehouse and doesn't need to be registered on any website. It's easy. It just requires dedicating space and registering. It shouldn't take more than a couple of days. 

Snowflake is reasonably priced, close to half the cost of some other solutions. 

In terms of performance the solution is good when compared to the analytical workloads and good in comparison to Redshift or BigQuery. The performance is on a slightly higher level, but when it comes to real-time performance, NoSQL is better than Snowflake, but that's in rare cases and depends on the particular requirement. Overall, for the analytical use case, Snowflake is a good solution and in terms of availability, it's a cloud data warehouse, so  they do replication and the like. 

It's important to understand your business needs, because these tools need to be properly modeled and they have their own advantages. If you're new to Snowflake, it's worth starting slowly for one month and move gradually, because if it's a complex system and you move everything to Snowflake without good architecture, then you can get stuck with the original problem. It's worth taking the time to make it efficient and then design modeling; there are SnowPro certifications as well. 

I rate this solution an eight out of 10.