INFORMATION_SCHEMA is BigQuery's built-in way to query metadata about your own BigQuery environment. Job history, table schemas, dataset listings, all exposed as views you can run SELECT against. The Associate Cloud Engineer exam tests this in a specific way, mostly around monitoring jobs and diagnosing usage issues. This article covers what INFORMATION_SCHEMA is, the views worth knowing, and the question patterns to watch for.
It does not cover every INFORMATION_SCHEMA view (there are many) or how to build elaborate cost-attribution dashboards on top of it. The goal here is what shows up on the ACE exam.
INFORMATION_SCHEMA is a set of read-only views that BigQuery automatically maintains. You query them with SQL, the same way you would query any other table. The data comes from BigQuery's own internal metadata, not from your tables.
Every project has its own INFORMATION_SCHEMA. Every dataset has its own as well. The views show you things like every job that ran, every table that exists, every column in every table, and how much data those queries scanned.
JOBS and JOBS_BY_PROJECT are the ones that come up most. They show recent job history. Who ran what query, when, how long it took, how much data it processed, and whether it succeeded or failed. JOBS_BY_USER and JOBS_BY_FOLDER and JOBS_BY_ORGANIZATION are variants that scope the same data differently. JOBS_BY_PROJECT is the most common one to query because most BigQuery analysis happens at the project level.
TABLES and COLUMNS show schema metadata. What tables exist in a dataset, what columns those tables have, what types those columns are. Useful for documenting a BigQuery environment or auditing what data you have.
SCHEMATA lists datasets within a project. Less commonly tested but worth knowing.
The most useful INFORMATION_SCHEMA query for the ACE exam framing is the one that shows recent jobs and how much data they processed:
SELECT
user_email,
job_id,
creation_time,
total_bytes_processed,
state
FROM
`my_project`.`region-us`.INFORMATION_SCHEMA.JOBS_BY_PROJECT
WHERE
creation_time > TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 24 HOUR)
ORDER BY
total_bytes_processed DESC
LIMIT 20That query gives you the 20 most expensive queries (by bytes processed) in the last 24 hours, who ran them, and what state they ended in. If a project is running over budget or hitting quota errors, this is one of the first places to look.
The region prefix matters. INFORMATION_SCHEMA views for jobs are scoped by region, so you query `region-us` or `region-eu` depending on where your jobs run.
INFORMATION_SCHEMA is BigQuery's native way to monitor BigQuery itself. Cloud Monitoring also collects BigQuery metrics and you can build dashboards there. The two are complementary.
Cloud Monitoring is better for real-time alerting, percentile latency on queries, slot utilization over time, and integrating BigQuery health into a broader Cloud Monitoring view across services. INFORMATION_SCHEMA is better for ad-hoc analysis with SQL, especially when you want to slice job history by user or query pattern.
If a question is about setting up an alert when query failures spike, that points to Cloud Monitoring. If a question is about figuring out which user ran the queries that blew through the daily quota yesterday, that points to INFORMATION_SCHEMA.
The Associate Cloud Engineer exam pattern is usually some variant of "how do I diagnose a BigQuery resource issue?" The exam specifically connect INFORMATION_SCHEMA to two scenarios.
First, monitoring recent jobs and resource usage. If a question asks how to see which BigQuery jobs ran recently and what they consumed, INFORMATION_SCHEMA is the answer.
Second, diagnosing a quotaExceeded error. The exam explicitly mention that INFORMATION_SCHEMA views are useful for understanding which jobs caused a quota to be hit. If a scenario describes a team hitting a BigQuery quota and trying to figure out why, the answer involves looking at INFORMATION_SCHEMA.JOBS_BY_PROJECT.
If you see "monitor recent BigQuery jobs", "diagnose quota exceeded", or "find expensive queries" in a question, think INFORMATION_SCHEMA. If the question is about real-time alerting on metrics, think Cloud Monitoring instead.
INFORMATION_SCHEMA is the SQL-queryable metadata layer inside BigQuery. JOBS_BY_PROJECT is the view to know for monitoring job activity and diagnosing usage issues. TABLES and COLUMNS are useful for schema-level questions. The Associate Cloud Engineer exam tests INFORMATION_SCHEMA most often in scenarios about diagnosing why a quota was hit or which queries are using the most resources.
My Associate Cloud Engineer course covers INFORMATION_SCHEMA in the BigQuery admin section alongside slot management, the BigQuery admin console, and how it complements Cloud Monitoring.
