Unclaimed: Are are working at Snowflake ?
Snowflake Reviews & Product Details
Snowflake is a cloud-based data warehousing platform designed to efficiently store, process, and analyze large volumes of data. It stands out for its ability to scale computing and storage resources independently, allowing users to manage data and analytics workloads more flexibly. With its unique architecture, Snowflake facilitates seamless data sharing among users, enabling collaborative data-driven decision-making. It supports various data structures and types, including structured and semi-structured data, making it a versatile choice for organizations looking to harness the power of their data.
| Capabilities |
|
|---|---|
| Ease of use |
|
| Deployment | Cloud / SaaS / Web-Based |
| Support | 24/7 (Live rep), Chat, Email/Help Desk, FAQs/Forum, Knowledge Base, Phone Support |
| Training | Documentation |
| Languages | English |
Snowflake Pros and Cons
- Multi-cloud support across AWS, Azure, and Google Cloud
- Separation of compute and storage for scalable, on-demand analytics
- Native support for semi-structured data formats like JSON and Avro
- Secure data sharing without moving or copying data
- Pay-per-second pricing model for cost optimization
- Advanced capabilities might be excessive for small businesses with limited data needs
- Potential for increased costs with high data storage and compute usage
- Learning curve for users unfamiliar with cloud-based data platforms
- Dependence on internet connectivity for accessing the platform
- Limited offline capabilities
Having the ability to spin up/down compute on demand is very useful. Traditional ETL patterns no longer need to be followed to get good results. Native JSON support makes dealing with APIs simple and easy.
Cumbersome security model is a pain to use but in reality much better for the data warehouse.
Having the ability to work with billions of rows of structured and unstructured data makes delivering insights much faster and easier.
Great computational power, back ups, ability to change size on demand.
The online interface is pretty good, but could use some improvement. I have experienced issues with complex queries that involve multiple joins and have a lot of partitions
Snowflake is used company wide by analysts and developers. Great performance and role management helped Snowflake's implementation
There are few things I liked Snowflake - -Ease of management -Complete elastic architecture -Complete separation of compute and storage -Compute clusters can be managed by end users -Simple pricing model -Dynamic scaling on the fly
I mostly liked Snowflake compared to other cloud databases and some aspects needs improvements incl. Geospatial functions and additional libraries.
I'm using Snowflake to house enterprise data platform where all critical insurance related end to end information is stored in the form of real-time dataware house platform. With Lambda-Snowpipe it has been greatly successful to solve business problems. End users are completely hooked to Snowflake functionality due to agility and usability. Since it enables self service cluster management it's even more popular.
Snowflake seamlessly blends into our BI stack, is invisible when we need it be, and is available and transparent when required.
The language is a little specific, and just different enough from regular SQL/similar.
Snowflake forms a seamless, consolidating layer between our BI layers, allowing complex analysis, previously very inefficient.
Performance and costs are configurable easily to your needs. Snowflake is also easily integrated with leading BI and ETL/ELT tools. Built-in tuning and optimization is very useful
Admin panel could be a bit more comprehensive (user/role/group management), usage metrics
Consolidated data warehouse and ability to handle data load, queries, and transactions simultaneously
- independent, on-demand scaling of compute and storage - no maintenance / ops (don't have to worry about vacuum, adding nodes, sort / dist keys, weekly maintenance, etc)
Because it's a new comer, the ecosystem is not as well established as Redshift (which has been around for over 6 years)
It serves as our data warehouse and it replaces Redshift. The biggest benefit is just how little time we have to spend on upkeep, and can invest that into building something that's much more impactful
The Snowflake capacity to deal with a really large amount of data, usually faster but if not you can always bump the warehouse size and get results faster, and this all is done within a few clicks and smooth. Very well documented as well and quick support answers when needed.
Lack of UI features for operations and debug. Not able to give hints to the compiler. Some updates that cause unexpected issues sometimes.
Basic ETL queries over a huge amount of data.
The performance has been great. We have seen 3 to 8 times improvement on query performance. There has been very low maintenance work such as tuning hardware or system configuration or query plan optimization. The auto resume and auto shutdown is great for cost savings. Also, we are using processing semi-structure data like JSON as well.
More improvements in on the Stored Proc front.
Migrating our on-premise MSSQL Server DW to Snowflake. We are improving performance, doing micro batching during the day, increase uptime (meeting SLAs), Centralizing our data and reducing complexity.
Snowflake lets me load a lot of formats of data and play around with them. It reads anywhere from .gz to .csv and let's me copy specific coloumns' data out. I actually use it as a replacement for Ms Excel on my Mac. I had taken a python course and instead of using Pandas, I used Snowflake to good effect.
Snowflake could bring in the support of a procedural SQL like PL/SQL. That would really make it unmatched.
We are using Snowflake for in house data warehousing and ETL. We run all our analytical and intelligence generation queries on Snowflake and have realized how fast and efficient it is.
Scalability - you can add as many compute nodes as you need to in order to meet your demand. You can also shut them down as needed to save on the cost.
Cost and perceived savings of auto shut off. While the concept is awesome, being able to auto shut off is the only way to make Snowflake come even close in terms of cost to Azure data warehouse and AWS redshift.
Servers as a data warehouse. Since it is a fully managed cloud service, a business doesn't need database administrators to set it up and perform management.