View Data Boost eligibility metrics

Bigtable Data Boost is a serverless compute service designed for high-throughput read jobs and queries. A workload must meet certain compatibility criteria to be able to use Data Boost.

This document describes how to check eligibility metrics to determine whether your workload is compatible with Data Boost. Review these metrics if you are are considering converting a standard app profile that is already in use to a Data Boost app profile.

Before you view eligibility metrics, review the details on ideal workloads and feature limitations at Bigtable Data Boost overview.

Eligibility metrics overview

To help you determine whether the workload for a given app profile is suitable for Data Boost, Bigtable provides two metrics:

  • Data Boost traffic eligibility count (data_boost/eligibility_count)
  • Data Boost traffic ineligible reasons (data_boost/ineligible_reasons)

For more information, see Metrics.

Check eligible and ineligible request counts

To determine whether traffic that is sent by a given app profile is suitable for Data Boost, look at the metric Data Boost traffic eligibility count. This metric shows the number of requests that are eligible or ineligible for Data Boost for a given app profile for the last seven days.

To view this metric for a specific app profile, follow these steps:

  1. Open the list of Bigtable instances in the Google Cloud console.

    Open the instance list

  2. Click the instance whose metrics you want to view.

  3. In the navigation pane, under Insights, click Monitoring.

  4. Click the Application profiles tab.

  5. View the Data Boost eligibility chart.

View an app profile's ineligibility reasons

The Data Boost traffic ineligible reasons metric tells you why requests sent with a given app profile are considered ineligible for Data Boost. A single request can have multiple reasons for being ineligible. Reasons for ineligibility include the following.

  • The app profile was used for requests that called API methods other than ReadRows, PingAndWarm, or SampleRowKeys.
  • The app profile was used for reverse scans.
  • Traffic was sent using an incompatible client library.
  • Requests using the app profile exceeded 1,000 read requests per second. For more information, see What it's not good for.

To view this metric for a specific app profile, follow these steps:

  1. Open the list of Bigtable instances in the Google Cloud console.

    Open the instance list

  2. Click the instance whose metrics you want to view.

  3. In the navigation pane, under Insights, click Monitoring.

  4. Click the Application profiles tab.

  5. View the Data Boost eligibility chart.

Conversion-blocking configurations

If you attempt to convert a standard app profile to use Data Boost, the conversion fails if any of the following conditions are true for the app profile.

  • The instance uses CMEK encryption.
  • It's configured for multi-cluster routing.
  • Single-row transactions are enabled.
  • It routes to a cluster in an instance that uses HDD storage.

What's next