[AMA CLOSED] Sisense/Snowflake Optimization Community AMA on March 22

[slosada]

Thank you to everyone that participated in our AMA on Sisense/Snowflake Optimization and a special thank you for our featured experts: 

Aaron Peabody

Co-Founder & CTO at Untitled Firm

Aaron specializes in the data element of Untitled’s business and leads the technical staff. His professional background has been predominantly comprised of technology startups and scale-ups, bringing an entrepreneurial edge and diverse tech stack to the Untitled team. Aaron has worked as a data strategist and integrations expert in a wide variety of industries. In former roles, he built data models for Fortune 100 companies to help guide decision-making for some of the largest marketing budgets in the world. His unique experience gives
him a large toolbox of skill sets to amplify a client’s data activation strategies.

Adriana Ontiveros Gonzalez

Jr. Data Scientist at Untitled Firm

Adriana studied Biomedical Engineering at the University of Monterrey and her Master’s Degree in Computer Science, as well as a Graduate Certificate in Data Science at the University of Louisville. Adriana is a data-driven individual and is passionate about Data Science and Data Analytics, and enjoys making data-based strategic recommendations for our partner’s projects.

This thread is now closed for further comment. 

[Community_Admin]

Why choose Snowflake for your Sisense Instance?

[untitled_Aaron]

What makes the Sisense/Snowflake relationship powerful is the ability to use either Sisense ElastiCubes or Live Models with any Snowflake instance. Allowing customers to use either ElastiCubes or Live Models allows for incredible flexibility for any use cases or SLAs that may exist.

[slosada]

Why choose Elasticube over Live models?

[untitled_Aaron]

In addition to supporting multiple data sources, ElastiCubes are a great option if there aren’t stringent data latency SLAs required from your data. If a refresh every hour doesn’t impact your business, then having an ElastiCube on a scheduled refresh may make a lot of sense.

[Community_Admin]

How Do Snowflake and Sisense Connect?

[Adriana_Onti]

When data is loaded into Snowflake, it is stored in an underlying database. To access or interact with that information, a cluster of compute resources is required. In Snowflake, each resource compute cluster is referred to as a virtual warehouse, or just simply, a warehouse. A warehouse is what Sisense will connect to as the engine to perform the necessary actions to support an ElastiCube or Live Model.

[slosada]

Why choose a Live Model over Elasticube?

[Adriana_Onti]

Live Models have one primary advantage over ElastiCubes: real-time data. If the nature of your business requires real-time data, or you have user SLAs that require real-time data, Live Models can support this.

[slosada]

How do I speed up my dashboard loading?

[untitled_Aaron]

The root of the issue lies with the location where Sisense is querying data from. The first level of optimization that should be considered is having a dedicated virtual warehouse for your BI tool. Then, Snowflake is a columnar data store, which is much faster than a row-store system. The major difference here being data is physically stored in columns rather than rows. There are several others that I would want to know more about specific issues to solve the issue!

[Community_Admin]

How do you create and assign security roles?

[Adriana_Onti]

All of the security role assignments can be written out using queries on the Snowflake UI. The security roles allow you to break down users into "groups" with different levels of access. Snowflake allows you to then assign these roles access as high as warehouse and databases down to row or column level access. The Snowflake documentation is a great place to start for this. It clearly lays out each permission for each level of access.

[slosada]

Are there use cases where you might be using more than 3 warehouses?

[Adriana_Onti]

Yes, for sure. We actually have a current client where we have put this into use. The 3 by 2 system works perfectly when everything is constrained to simply moving data from the source, transforming for production, and loading into Sisense. The use case for a fourth warehouse comes in handy when you have "too many hands in the cookie jar". For instance, this client has an entire research team that is actively using Snowflake daily to improve their data quality. This team was previously using the PROD_WAREHOUSE (meant only to move data from production database to Sisense). This was causing a conflict over the compute resource and a jump in Snowflake costs. We fixed this by adding a RESEARCH_WAREHOUSE. This fourth warehouse allowed us to better understand how compute resources were being used, where costs were coming from, and allowed Sisense dashboards to not be impacted by backend queries being written on the same warehouse.

[Community_Admin]

What is Snowflake access control in a nutshell?

[untitled_Aaron]

Access control is similar to most databases and warehouses where a user is created to access Snowflake. Users can be granted permissions directly or have roles assigned to them with permissions already granted. Users can only view use the databases and warehouses they have permissions to and perform actions granted to them. You can authenticate using username & password, federation authentication & sso, key pairs, oauth and you can also implement MFA.