3 Slingshot updates for simplified warehouse management
The Capital One Slingshot team has been busy with a string of updates this spring. These include three exciting updates to enhance a customer’s experience with Slingshot:
- The ability to view and manage Snowflake Query Acceleration Service (QAS) at a warehouse level from Slingshot
- A new all-in-one Recommendations page
- A homepage refresh to feature the most actionable insights front and center after logging into Slingshot. Let’s look at these updates and how they help customers maximize value from Slingshot.
Managing Query Acceleration Service (QAS) in Slingshot
Slingshot now allows users to easily identify and manage QAS enabled warehouses.
Note: Read more about best practices on using QAS with Slingshot in Slingshot documentation.
Identifying QAS enabled warehouses in Slingshot
Within Slingshot, there are a number of pages with the warehouse detail view with a new column indicating whether QAS has been enabled or disabled. Below are a few examples of where you can find this new QAS attribute:
- Under the “Your Warehouses” list, the new QAS column shows which warehouses have QAS enabled or disabled.
- Within the “Warehouse Details” view under “Approvals Inbox” and “Recommendations” we can see the QAS attribute as enabled or disabled, with the scale factor value for QAS enabled warehouses also displayed.
Creating and managing QAS enabled warehouses in Slingshot
Slingshot already empowers certain users to provision new warehouses with schedules to dynamically change warehouse sizes and configuration based on workload needs. In a continued effort to make Slingshot a one-stop-shop for provisioning and managing warehouses, users can now enable or disable QAS, as well as set the scale factor within Slingshot, for new and existing warehouses. QAS settings can be found and modified during warehouse creation and modification flows, as follows:
- Under “Requests”, “Create New Warehouse” will let users enable QAS and add a corresponding scale factor.
- Under “Your Warehouses”, the “Modify” action allows enabling or disabling QAS and changing the scale factor for a QAS enabled warehouse.
Understanding the scale factor in relation to Slingshot schedules
Slingshot’s scheduling tools empower users to dynamically change warehouse size and other parameters by hour and day of the week. QAS scale factor then can act as a multiplier of computing resources based on warehouse size and credits and sets an upper bound on the amount of compute a warehouse can additionally lease for query acceleration execution. QAS’s scale factor has a default setting (8 credits) that is based on the warehouse’s existing schedule block.
Slingshot recommendations made simple
Slingshot regularly analyzes warehouse usage patterns to generate recommendations for right-sizing warehouses within a defined schedule on a daily basis (detailed documentation here). Slingshot has a new and improved Recommendations page to increase user confidence when applying recommendations. The new Recommendation page, features an all-in-one view page with some new enhancements over the previous detailed view:
New summary section
The summary section provides a quick explanation of why Slingshot recommends this optimization for this specific warehouse, highlighting corresponding issues, along with current cost and performance as well as projected result upon applying this recommendation.
An updated schedule comparison
The updated schedule comparison allows users to toggle between the current and recommended schedules to see which time blocks need attention, based on Slingshot’s hourly analysis of a given warehouse’s performance. The new current schedule will also flag areas of opportunity in red. Selecting any red time block displays the corresponding issue list and explanation, as analyzed by Slingshot.
Query performance impact analysis
The new Recommendation page features a new powerful tool to gain insights about the impacts of the identified issues and how taking Slingshot’s suggestions improves performance or costs - “Projected impact on queries”.
This new tool provides visibility into specific query performance impact before and after applying recommendations to a warehouse.
Updated Slingshot Homepage
Thanks to invaluable feedback from our customers, we've moved some of the most impactful features in Slingshot front and center, right when user’s log in:
- A new top recommendations list
- A refined ‘costliest queries’ list
- A handy upfront list of pending warehouse approval requests.
Your top recommendations, now front and center
We’ve added insights about “Your Top Recommendations” to the Homepage, bringing the most impactful optimization opportunities front and center for users, immediately upon logging into Slingshot. Users can toggle between cost or performance recommendations, see how much cost reduction or how much average query execution time can be improved by applying these recommendations. Users can also navigate to all Slingshot recommendations right from this area.
Costly Queries
Slingshot has also added the Costly Queries dashboard to the new Homepage experience. Users can see the execution time, frequency and the cost of each query, as well as the query ID for identifying the exact query in Snowflake. This helps with identifying and eliminating any wasteful or inefficient queries driving up Snowflake costs.
Besides identifying costliest queries, you can use Slingshot’s Query Advisor to analyze and improve query performance.
Pending approval requests
With the approval workflow feature enabled in Slingshot, pending approval requests will now also live on the homepage. These requests can quickly be reviewed, along with the details of each request, such as the requester, the request type and max monthly cost with a button allowing for a decision to approve or decline.
We wanted to bring these features and updates to the homepage of Slingshot so that users can get the most out of it as soon as they log in. Users can now more quickly enable greater savings and gain the confidence that their data environment is running smoothly and efficiently after onboarding Slingshot to their Snowflake data cloud.