# Benchmarks

> Tested on v1.0.0. All tests passed (37/37).

## CRUD — Small (100 rows)

| Operation                         | Time   | Throughput     |
| --------------------------------- | ------ | -------------- |
| INSERT sequential (100 rows)      | 72 ms  | 1,388 rows/s   |
| INSERT via TRANSACTION (100 rows) | 30 ms  | 3,334 rows/s   |
| SELECT sequential (200 queries)   | 107 ms | 1,870 rows/s   |
| UPDATE all 100 rows               | 1 ms   | 100,000 rows/s |

## CRUD — Medium (1,000 rows)

| Operation                             | Time  | Throughput     |
| ------------------------------------- | ----- | -------------- |
| BATCH INSERT via prepare (1,000 rows) | 17 ms | 58,824 rows/s  |
| SELECT ALL (1,000 rows)               | 12 ms | 83,334 rows/s  |
| SELECT WHERE + ORDER BY (\~100 rows)  | 2 ms  | 50,000 rows/s  |
| UPDATE all 1,000 rows                 | 5 ms  | 200,000 rows/s |
| INSERT via TRANSACTION (500 rows)     | 98 ms | 5,102 rows/s   |

## CRUD — Large (10,000 rows)

| Operation                              | Time  | Throughput     |
| -------------------------------------- | ----- | -------------- |
| BATCH INSERT via prepare (10,000 rows) | 82 ms | 121,952 rows/s |
| SELECT ALL (10,000 rows)               | 44 ms | 227,272 rows/s |
| SELECT WHERE + ORDER BY (\~1,000 rows) | 5 ms  | 200,000 rows/s |
| UPDATE all 10,000 rows                 | 36 ms | 277,778 rows/s |

## JSONB — By Payload Size

| Payload | Operation     | Rows   | Time   | Throughput     |
| ------- | ------------- | ------ | ------ | -------------- |
| 256 KB  | WRITE         | 792    | 26 ms  | 30,462 rows/s  |
| 256 KB  | READ ALL      | 792    | 11 ms  | 72,000 rows/s  |
| 256 KB  | FIELD QUERY   | —      | 1 ms   | —              |
| 256 KB  | EXTRACT field | 792    | 3 ms   | 264,000 rows/s |
| 1 MB    | WRITE         | 3,168  | 93 ms  | 34,064 rows/s  |
| 1 MB    | READ ALL      | 3,168  | 41 ms  | 77,268 rows/s  |
| 1 MB    | FIELD QUERY   | —      | 2 ms   | —              |
| 1 MB    | EXTRACT field | 3,168  | 5 ms   | 633,600 rows/s |
| 5 MB    | WRITE         | 15,840 | 536 ms | 29,552 rows/s  |
| 5 MB    | READ ALL      | 15,840 | 148 ms | 107,028 rows/s |
| 5 MB    | FIELD QUERY   | —      | 5 ms   | —              |
| 5 MB    | EXTRACT field | 15,840 | 26 ms  | 609,230 rows/s |

## Highlights

* **Fastest write**: BATCH INSERT via prepare — up to **121,952 rows/s** at 10k rows
* **Fastest read**: SELECT ALL at 10k — **227,272 rows/s**
* **Fastest update**: UPDATE at 10k — **277,778 rows/s**
* **JSONB extract** scales extremely well — **633,600 rows/s** on 1 MB payload
* **TRANSACTION overhead** is significant at small scales (3,334 vs 58,824 rows/s for batch)


---

# Agent Instructions: 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:

```
GET https://prysma-studio.gitbook.io/pry-pgadapter/overview/benchmarks.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.