Hardware-accelerated CRC32 based on Intel Streaming SIMD Extensions 4.2


Starting with the Nehalam series, Intel processors feature the Streaming SIMD Extensions instruction set which provide a hardware-accelerated version of the CRC32 algorithm (Castagnoli variant). This library uses the Intel SSE 4.2 instruction set to provide a fast CRC-32 algorithm.

  • Intel Streaming SIMD Extensions 4.2 based hardware accelerated CRC 32 calculation
  • Progressive CRC32 calculation
  • Supports Node.js buffers

The tests were run on a Macbook Air running an Intel Core i7 processor, with 8GB of RAM and used buffers instead of strings to prevent having items on the V8 heap that might cause the garbage collector to fire frequently and interfere with the test run-times.

Below are the results from the 2 test cases:

>node benchmark/1.single_1kb_length_buffer.benchmark.js
100000 calls to calculate CRC on a 1024 byte buffer...
        SSE4.2 based CRC32: 26ms.
        Pure JS based CRC32 (table-based): 699ms.
        Pure JS based CRC32 (direct): 3704ms.

>node benchmark/2.multi_random_length_buffer.benchmark.js
100000 calls to calculate CRC on random length buffers upto 4096 bytes long...
        Avg. buffer length: 2042 bytes
        SSE4.2 based CRC32: 62ms.
        Pure JS based CRC32 (table-based): 1968ms.
        Pure JS based CRC32 (direct): 8220ms.

The results clearly show that the SSE4_CRC32 library is about 31.74 times faster than the pure JS library!

Use the following command to install the library from npm:

npm install sse4_crc32

Using the library is quite simple. Start out by including the library in your code:

var SSE4CRC32 = require("sse4_crc32");

To calculate the 32-bit CRC for any string, simply use the following code:

var crc = SSE4CRC32.calculate("my string");

Instead of passing in a string, a buffer can be passed to the calculate() function.

To calculate CRC in a progressive manner, use the code snippet:

var sse4crc32 = new SSE4CRC32.CRC32(),
    my_inputs = [ "some string", new Buffer("a buffer"), "yet another string", new Buffer("yet another buffer") ],

my_inputs.forEach(function (string) {
    crc = sse4crc32.update(string);

crc = sse4crc32.crc();              // The .crc() method can also be used to get the final CRC

Also see the example code in the examples directory.

Once the repository has been cloned, use one of the following commands to build the library:

make all            // Builds the release version of the library and runs all tests
make debug          // Builds the debug version of the library
make clean          // Removes all files generated by builds

All feedback/suggestions/criticisms can be directed to Anand Suresh