> For the complete documentation index, see [llms.txt](https://mariadb.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://mariadb.com/docs/analytics/mariadb-columnstore/high-availability/optimization-and-tuning/analyzing-queries-in-columnstore.md).

# Analyzing Queries in ColumnStore

The `calGetStats()` function provides execution resource statistics for the most recent query run within the current user session. Because MariaDB ColumnStore leverages a distributed columnar architecture instead of traditional row-based indexes, evaluating these metrics helps database administrators verify the efficiency of table scans, memory cache hits, and filter performance.

## Core Query Execution Metrics

When analyzing query performance, three primary tracking values indicate how efficiently data is being isolated across the processing layers:

### Physical I/O (`PIO` / `ApproxPhyI/O`)

* Definition: The total count of 8 KB blocks read directly from physical disk subsystems or external cloud storage to satisfy the query execution plan.
* Analysis Context: High values represent a cold cache state where data blocks must be fetched from persistent storage. Successive executions of identical or overlapping query workloads should result in this metric dropping to `0` as the blocks populate the local performance memory cache.

### Logical I/O (`LIO` / `BlocksTouched`)

* Definition: The total number of logical 8 KB blocks processed during query execution, also formally known as Blocks Touched.
* Analysis Context: This value captures every block examined by the engine primitives—including blocks retrieved from memory cache pools as well as those read from disk. A disproportionately high count relative to the number of rows returned indicates that the query is evaluating wide data segments to isolate its results.

### Partition Blocks Eliminated (`PBE` / `PartitionBlocksEliminated`)

* Definition: The number of 8 KB column blocks successfully bypassed by the query processor using the metadata value ranges tracked inside the Extent Map.
* Analysis Context: ColumnStore registers minimum and maximum thresholds for every data extent to skip chunks that cannot contain records matching your `WHERE` filter parameters. A high block elimination count indicates highly optimized query targeting. If this value registers as `0` for a heavily filtered query, the engine was forced to scan every block for that column, implying the data on disk is not logically ordered or grouped by that filter criteria.

## Query Analysis Sequence

Execute the following steps within an active database session to measure baseline query I/O and verify if block elimination is occurring properly:

1. Clear the performance cache (Optional for cold-run simulation):

   ```sql
   SELECT calFlushCache();
   ```
2. Run the target query statement:

   ```sql
   SELECT COUNT(*), region_id FROM orders WHERE region_id = 4 GROUP BY region_id;
   ```
3. Retrieve the metrics summary for the execution:

   ```sql
   SELECT calGetStats();
   ```

   *Example Output String:*

   Plaintext

   ```
   Query Stats: MaxMemPct-0;NumTempFiles-0;TempFileSpace-0B;ApproxPhyI/O-1931;CacheI/O-2446;BlocksTouched-2443;PartitionBlocksEliminated-48;MsgBytesIn-73KB;MsgBytesOut-1KB;Mode-Distributed
   ```


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://mariadb.com/docs/analytics/mariadb-columnstore/high-availability/optimization-and-tuning/analyzing-queries-in-columnstore.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.