+++
title = "Step 3 - Scan the employees table to find managers by using the sparse global secondary index"
date = 2019-12-02T10:57:24-08:00
weight = 3
+++


Now, scan the new global secondary index `GSI_2` on the `employees` table.  In using our new sparse index we expect that we'll consume read capacity for fewer items.  We'll use the sparse index as a very effective filter to improve efficiency for this access pattern.

```py
response = table.scan(
  Limit=pageSize,
  IndexName='GSI_2'
)
```
Run the following AWS CLI command to execute this scan using the sparse index.
```bash
python scan_for_managers_gsi.py employees 100
```
**Parameters:**
1. Table name = `employees`
1. Page size = `100` (this is size of the pagination for the scan).

The following output includes the scanned count and the execution time.
```txt
Number of managers: 84. # of records scanned: 84. Execution time: 0.287754058838 seconds
```

Observe the scanned count and execution time using the sparse index.  How does this compare to the result achieved from the Scan of the base table in Step 2? The sparse index has less data and is more efficient.