# Store application data

Create records, query them, and group related changes in transactions.

Source: https://particledb.ai/guides/relational-data/

This guide uses a small work queue stored as an ordinary SQL table. Run the SQL in a connected SQL client or the SQL editor.

## Create and populate a table

```sql
CREATE TABLE work_items (
  id BIGINT PRIMARY KEY,
  title TEXT NOT NULL,
  status TEXT NOT NULL
);
INSERT INTO work_items VALUES
  (1, 'Review request', 'pending'),
  (2, 'Send response', 'pending');
```

## Read only the records you need

```sql
SELECT id, title
FROM work_items
WHERE status = 'pending'
ORDER BY id;
```

Use bound parameters for values supplied by users. The placeholder syntax depends on the client; see your [language guide](/sdks/overview/).

## Commit related changes together

Use one connection for the transaction:

```sql
BEGIN;
UPDATE work_items SET status = 'completed' WHERE id = 1;
UPDATE work_items SET status = 'running' WHERE id = 2;
COMMIT;
```

If a statement fails, roll back before reusing the connection. Retry from `BEGIN` when a transaction fails with a retryable conflict.

## Check the result

```sql
SELECT id, status FROM work_items ORDER BY id;
```

The statuses are `completed` for record 1 and `running` for record 2.

## Next steps

[Transactions](/sql/transactions/) · [Data types](/sql/data-types/) · [Indexes and queries](/guides/query-performance/)