Speed up slow-loading dashboard charts by enabling materialized-view-based acceleration. Simple Log Service (SLS) pre-computes query results in the background so that dashboards render in seconds instead of tens of seconds.
How it works
When SQL queries in a dashboard involve large data volumes or long time spans, chart rendering can take tens of seconds or time out entirely. Dashboard query acceleration addresses this by creating materialized views that continuously pre-compute and persist query results. At query time, SLS reads the pre-computed results instead of re-running the original complex queries.
Prerequisites
-
An SLS project with at least one dashboard that contains SQL-based charts.
-
Charts with SQL execution latency of 5 seconds or more benefit most from acceleration.
Procedure
-
Log on to the SLS console.
-
Open the target project. In the left-side navigation pane, choose Dashboard > Dashboards.
-
Open the target dashboard and click the Materialized View Acceleration button. SLS analyzes the execution performance of every chart in the dashboard.
-
Smart detection: By default, SLS identifies charts that support materialized view acceleration and have SQL execution latency greater than 5 seconds.
-
Custom threshold: Adjust the latency threshold to match your performance requirements, and then click Re-analyze.
-
Precise selection: Review the suggested materialized view names and their corresponding charts. Select the checkbox for each chart to accelerate.
-
-
Click Next. Configure the materialized view parameters and save the configuration.
Parameter
Description
Default
Recommendation
Acceleration start time
The point in time from which SLS backfills pre-computed data.
One week ago
For charts that use long-term comparisons such as year-over-year analysis, set an earlier start time to ensure full data coverage.
Update interval
The frequency at which SLS re-runs the pre-computation.
15 minutes
For very large data volumes, increase the interval to reduce compute resource consumption.
What to expect
After you save the configuration, SLS creates the materialized view in the background and begins asynchronous pre-computation of historical data. The backfill duration depends on the volume of historical data. After the pre-computation covers the current query range, dashboard charts load in seconds.