Benchmark: msgpackr vs. JSON v2

HTML Preparation code:

<script src="https://unpkg.com/msgpackr@1.11.2/dist/index.min.js"></script>

Script Preparation code:

var stringData = {"glossary":{"title":"example glossary","GlossDiv":{"title":"S","GlossList":{"GlossEntry":{"ID":"SGML","SortAs":"SGML","GlossTerm":"Standard Generalized Markup Language","Acronym":"SGML","Abbrev":"ISO 8879:1986","GlossDef":{"para":"A meta-markup language, used to create markup languages such as DocBook.","GlossSeeAlso":["GML","XML"]},"GlossSee":"markup"}}}}};
var numberData = [
  [1, 2, 3, 4, -1, true, null],
  [3, 6, 5, 4,  1, false, 7],
  [3, 2, 8, 1,  0, true, 0],
  [10, 11, 12, 13,  14, false, true],
  [15, 16, 17, 18, 19, true, null],
  [20, 21, 22, 23,  24, false, 7],
  [25, 26, 27, 28,  29, true, 0],
  100, 200, 300,
  [
    [1, 0, 0, 0, 0],
    [0, 1, 0, 0, 0],
    [0, 0, 1, 0, 0],
    [0, 0, 0, 1, 0],
    [0, 0, 0, 0, 1]
  ]
];
var jsonEncodedStrings = JSON.parse(JSON.stringify(stringData)),
    jsonEncodedNumbers = JSON.parse(JSON.stringify(numberData)),
    msgpackEncodedStrings = msgpackr.unpack(msgpackr.pack(stringData)),
    msgpackEncodedNumbers = msgpack.unpack(msgpackr.pack(numberData));

Tests:

MsgPack Strings Encode
var result = msgpackr.unpack(msgpackr.pack(stringData));
JSON Strings Encode
var result = JSON.parse(JSON.stringify(stringData));
MsgPack Numbers Encode
var result = msgpackr.unpack(msgpackr.pack(numberData));
JSON Numbers Encode
var result = JSON.parse(JSON.stringify(numberData));

Rendered benchmark preparation results:

Suite status: <idle, ready to run>

Previous results

Test case name	Result
MsgPack Strings Encode
JSON Strings Encode
MsgPack Numbers Encode
JSON Numbers Encode

Fastest: N/A

Slowest: N/A

Latest run results:

Run details: (Test run date: one month ago)

User agent: Mozilla/5.0 (iPhone; CPU iPhone OS 18_3_1 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.3 Mobile/15E148 Safari/604.1

Browser/OS: Mobile Safari 18 on iOS 18.3.1

View result in a separate tab

Test name	Executions per second
MsgPack Strings Encode	354226.4 Ops/sec
JSON Strings Encode	1341442.4 Ops/sec
MsgPack Numbers Encode	626854.6 Ops/sec
JSON Numbers Encode	595172.0 Ops/sec

Autogenerated LLM Summary (model gpt-4o-mini, generated 3 months ago):

LLMs can make mistakes. Check important info.

The benchmark you provided compares the performance of two data serialization formats: **MsgPack** (using the `msgpackr` library) and **JSON**. The focus is on encoding operations for both string and number data structures, with the benchmark designed to determine which format's encoding yields better performance.

### Options Compared

1. **MsgPack vs JSON for Strings Encoding**
   - **Benchmark Definition**: 
     - MsgPack: `msgpackr.unpack(msgpackr.pack(stringData));`
     - JSON: `JSON.parse(JSON.stringify(stringData));`
   - **Test Name**: "MsgPack Strings Encode" vs "JSON Strings Encode"

2. **MsgPack vs JSON for Numbers Encoding**
   - **Benchmark Definition**: 
     - MsgPack: `msgpackr.unpack(msgpackr.pack(numberData));`
     - JSON: `JSON.parse(JSON.stringify(numberData));`
   - **Test Name**: "MsgPack Numbers Encode" vs "JSON Numbers Encode"

### Libraries and Syntax

- **MsgPackr**: 
  - This is a JavaScript library that implements the MessagePack serialization format. MessagePack is a binary format that is efficient for storage and transfer. It is designed to be compact, providing smaller payloads compared to JSON, while still being fast to serialize and unserialize. The usage of `msgpackr` in the benchmark demonstrates how well it can handle different data structures compared to traditional JSON encoding.

- **JSON**:
  - JSON (JavaScript Object Notation) is a lightweight data interchange format easy for humans to read and write and for machines to parse and generate. The benchmark uses `JSON.parse` and `JSON.stringify` to encode and decode the data structures. Although JSON is widely supported, it is a text-based format that can result in larger data sizes compared to binary formats like MessagePack.

### Performance Results

The benchmark results show the number of executions per second for each test case on the given mobile platform (Chrome Mobile 132 on Android). Here's a summary:

- **JSON Strings Encode**: 327,308.03125 executions/second (fastest)
- **MsgPack Numbers Encode**: 308,838.40625 executions/second
- **JSON Numbers Encode**: 279,994.15625 executions/second
- **MsgPack Strings Encode**: 73,567.5234375 executions/second (slowest)

### Pros and Cons

**MsgPack Pros**:
- More efficient serialization for binary data, leading to smaller payloads.
- Faster for certain data types compared to JSON.

**MsgPack Cons**:
- Not human-readable; debugging is more complex.
- Slightly larger initialization overhead due to the need for unpacking.

**JSON Pros**:
- Human-readable format that is easy to debug.
- Universally supported across programming languages.

**JSON Cons**:
- Larger data size compared to MsgPack, leading to higher storage and transmission costs.
- Can be slower for parsing complex data structures.

### Alternatives

In addition to MsgPack and JSON, other serialization formats can be considered:

1. **Protocol Buffers**: Developed by Google, offers high efficiency and is suitable for large-scale applications. However, it requires a predefined schema.
2. **Avro**: Works particularly well with large data sets, dynamic typing, and supports schema evolution.
3. **CBOR** (Concise Binary Object Representation): Like MessagePack but can offer even more features, including richer data types.
4. **XML**: While more verbose than JSON and MsgPack, it is still used in many legacy systems and provides extensive metadata capabilities.

In summary, this benchmark highlights the performance differences between MsgPack and JSON in encoding string and numeric data, providing insights into when and why one might choose a serialization method over the other.

Related benchmarks:

msgpackr vs. JSON v2 (version: 1)

Comparing performance of: MsgPack Strings Encode vs JSON Strings Encode vs MsgPack Numbers Encode vs JSON Numbers Encode

Created: 3 months ago by: Guest

Jump to the latest result

MsgPack Strings Encode

JSON Strings Encode

MsgPack Numbers Encode

JSON Numbers Encode

Suite status: <idle, ready to run>

Experimental features:

Fastest: N/A

Slowest: N/A

Autogenerated LLM Summary (model gpt-4o-mini, generated 3 months ago):