By: Joseph Chapman, Group Product Manager
As part of Analytics Builder, your organization is entitled to an allocation of an in-memory query and storage engine called SPICE. As your data and reporting needs grow over time, you may need to reduce or reallocate your SPICE usage to stay within these limits. This article helps you decide whether SPICE is right for your data needs, details how to view your organization’s total SPICE allocation, explains how to proactively monitor usage over time with automatic alerts, and outlines actions you can take to reduce SPICE usage, including converting existing datasets to Direct Query mode, optimizing the amount of SPICE a dataset consumes, and removing SPICE datasets that are no longer needed.
When to Use SPICE vs. Direct Query
Depending on the use case, authors can leverage either SPICE or Direct Query mode with their datasets. The following table outlines key considerations and constraints for each query mode.
Attribute | SPICE (Super-fast, Parallel, In-memory Calculation Engine) | Direct Query |
|---|
How it works | Imports and caches data directly into in-memory storage. | Sends live SQL queries directly to the underlying data source whenever a user opens or interacts with a dashboard. |
Data freshness | Stores a snapshot of data in-memory, enabling fast dashboard loads, report generation, and calculations. Data freshness depends on scheduled, manual, or API-triggered refreshes, so data may not always reflect the latest source changes. | Stores a snapshot of data in-memory, enabling fast dashboard loads, report generation, and calculations. Data freshness depends on scheduled, manual, or API-triggered refreshes, so data may not always reflect the latest source changes. |
Performance | Delivers ultra-fast dashboard load times and high-speed advanced calculations. | Depends entirely on the speed and performance of the backend system. |
Capacity limits | Supports up to 2 billion rows per dataset. Capacity is limited to your organization's entitled SPICE allocation often in gigabytes. | No strict imposed row or size limits, making it suitable for datasets that exceed your organization’s SPICE limits. |
Cost & load | Reduces query load on the source database because queries occur only during refreshes, lowering downstream query costs. | Places continuous query load on the source data during user interactions. |
External data sources | Required for uploaded files and other non-queryable external data sources. | Not supported for uploaded files or non-queryable external data sources. |
Data transformations | Supports dataset transformations including: Pivot, Unpivot, Aggregate, Append, and more complex multi-step or multi-branch data preparation and transformation workflows. | Limited transformation support. Does not support Pivot, Unpivot, Aggregate, Append, or complex/multi-branch data preparation workflows, as queries must be executed directly against the source system. |
Now that you understand the considerations and trade-offs between SPICE and Direct Query, the following sections outline the tools and steps available in Analytics Builder to help you monitor, manage, and optimize your organization's SPICE usage.
What's New in October 2026
Starting in the October 2026 release of Brightspace, users will be able to view their organization’s SPICE allocation and usage details using a new pre-built dashboard.
To view the new SPICE Capacity Usage dashboard:
- Log in to Analytics Builder as an author
- Click on Dashboards
- Search for "SPICE Capacity Usage"
The SPICE Capacity Usage dashboard provides essential information including your organization's current SPICE allocation, how much SPICE has been consumed, how much remains, how many datasets are leveraging SPICE, which datasets are using SPICE, how much each dataset is consuming, who authored the dataset, and when this report was last refreshed. There are also various filtering capabilities to help you find and drill down into the various SPICE datasets.
The SPICE Capacity Usage dashboard refreshes daily. The dashboard also shows all SPICE datasets for an organization including datasets not created or owned by the logged in author.
Setting Automatic SPICE Alerts
Authors can create alerts to be notified when SPICE usage reaches certain levels. To create an alert:
- Visit the SPICE Capacity Usage dashboard
- Hover over a KPI metric to monitor (e.g., “Remaining”)
- Click Create alert
- Enter a condition and threshold (e.g., “Is below 0”)
- Press Save
To manage existing alerts, click the Alerts button from the dashboard’s toolbar.
Note: After performing any of the actions below, changes to your SPICE usage do not immediately appear in the SPICE Capacity Usage dashboard. Allow up to a day for updated usage to be reflected.
Converting from SPICE to Direct Query
Consider converting an existing SPICE dataset to use Direct Query mode when the dataset is still actively used in dashboards, analyses, or reports, but the underlying data doesn't require SPICE's in-memory acceleration. Not all SPICE datasets can be converted to Direct Query; refer to the external data sources and data transformation details in the table above. For eligible datasets, converting query modes provides a direct path to reduce SPICE consumption for low-concurrency dashboards, infrequently changing data, datasets built entirely from existing Brightspace data sources with no new or external data transformations, or sources that are already optimized for query performance.
Authors can convert their datasets from SPICE to Direct Query mode using the following steps:
- Visit the SPICE Usage Dashboard in Analytics Builder
- Review the list of datasets and identify one that is using SPICE
- Copy or note the dataset name
- Navigate to Datasets
- Search for the dataset name
- Click Edit from the details icon (⋮)
If using the new data prep experience:
- Click the SPICE dropdown located in the upper right of the screen
- Choose Direct Query
- Click Save & publish
If using the legacy data prep experience:
- Locate Query mode in the lower left of the screen
- Select the Direct query radio option
- Click Save & publish
Reducing SPICE Consumption per Dataset
If converting a dataset from SPICE to Direct Query is not possible, authors may be able to reduce the amount of SPICE being consumed by optimizing their data. Consider the following techniques:
- Exclude or remove columns that are not required by analyses, dashboards, or filters
- Apply filters to limit the rows imported in SPICE (e.g., applying date-range filters)
- Remove unused calculated fields that are materialized in the dataset
- Convert fields stored as strings into dates, integers, or decimals where appropriate
- Remove any unnecessary calculated fields that can be moved into visuals or that are no longer needed
- If possible, reduce data granularity before importing external data into Analytics Builder (e.g., convert transaction-level data into daily/monthly summaries)
Deleting a SPICE Dataset
Consider removing an existing SPICE dataset when it is no longer referenced by any active dashboard or analysis. This is often the case for datasets created for a one-time or time-boxed reporting need that has since concluded, or for datasets that duplicate data already available through another dataset in use.
To view if a SPICE dataset is being used elsewhere:
- Visit the SPICE Usage Dashboard in Analytics Builder
- Review the list of datasets and identify one that is using SPICE
- Copy or note the dataset name
- Navigate to Datasets
- Search for the dataset name and click on it
- View the Usage information in the lower corner
To delete a dataset:
- Navigate to the Datasets page in Analytics Builder
- Click the details icon (⋮)
- Click Delete
- Review the confirmation dialog which outlines any affected users and usage
- Click Delete