Go v1.10.3 vs Node v10.6.0 Microbenchmark

This post is making use of the same code from a previous benchmark: Go v1.9.1 vs Node v8.9.0 Microbenchmark. Check that out for some background and code samples if you'd like to try the same experiment. The applications are very simple. There is a global counter which increments with each incoming HTTP request. Both applications reply with the same data format, using simple string concatenation, and no third party libraries.

Here's the command we're using to execute the benchmark:

siege -t 2m -b http://localhost:8000/

The results are as follows:

Node v10.6.0

Transactions: 183506 hits
Availability: 100.00 %
Elapsed time: 119.15 secs
Data transferred: 2.52 MB
Response time: 0.01 secs
Transaction rate: 1540.13 trans/sec <--
Throughput: 0.02 MB/sec
Concurrency: 14.97
Successful transactions: 183506
Failed transactions: 0
Longest transaction: 0.07
Shortest transaction: 0.00

Go v1.10.3

Transactions: 368668 hits
Availability: 100.00 %
Elapsed time: 119.83 secs
Data transferred: 4.81 MB
Response time: 0.00 secs
Transaction rate: 3076.59 trans/sec <--
Throughput: 0.04 MB/sec
Concurrency: 14.95
Successful transactions: 368668
Failed transactions: 0
Longest transaction: 0.07
Shortest transaction: 0.00

Takeaway

Go is 2.00x as fast as Node.js for this particular benchmark. It's worth noting that these Node.js results are the same as they were for Node v8. However the Go v1.10 results are about 20% faster than with v1.9.

Thomas has contributed to dozens of enterprise Node.js services and has worked for a company dedicated to securing Node.js. He has spoken at several conferences on Node.js and JavaScript and is an O'Reilly published author.