github.com/audriusbutkevicius/go_serialization_benchmarks

Benchmarks of Go serialization methods

This is a test suite for benchmarking various Go serialization methods.

Tested serialization methods

• encoding/gob
• encoding/json
• github.com/alecthomas/binary
• github.com/davecgh/go-xdr/xdr
• github.com/Sereal/Sereal/Go/sereal
• github.com/ugorji/go/codec
• gopkg.in/vmihailenco/msgpack.v2
• labix.org/v2/mgo/bson
• github.com/tinylib/msgp (code generator for msgpack)
• github.com/golang/protobuf (generated code)
• github.com/gogo/protobuf (generated code, optimized version of goprotobuf)
• github.com/DeDiS/protobuf (reflection based)
• github.com/google/flatbuffers
• github.com/hprose/hprose-go/io
• github.com/glycerine/go-capnproto
• zombiezen.com/go/capnproto2
• github.com/andyleap/gencode
• github.com/pascaldekloe/colfer

Running the benchmarks

go get -u -t
go test -bench='.*' ./

Shameless plug: I use pawk to format the table:

go test -bench='.*' ./ | pawk -F'\t' '"%-40s %10s %10s %s %s" % f'

Recommendation

If performance, correctness and interoperability are the most important factors, gogoprotobuf is currently the best choice. It does require a pre-processing step (eg. via Go 1.4's "go generate" command).

But as always, make your own choice based on your requirements.

Data

The data being serialized is the following structure with randomly generated values:

type A struct {
    Name     string
    BirthDay time.Time
    Phone    string
    Siblings int
    Spouse   bool
    Money    float64
}

Results

Results with Go 1.6 on a 2.6 GHz Intel Core i7 (MacBook Pro Retina 15-inch, late 2013):

benchmark                                   iter           time/iter      bytes alloc               allocs
---------                                   ----           ---------      -----------               ------
BenchmarkMsgpMarshal-8                      5000000        269 ns/op      128 B/op        1 allocs/op
BenchmarkMsgpUnmarshal-8                    3000000        498 ns/op      112 B/op        3 allocs/op
BenchmarkVmihailencoMsgpackMarshal-8        1000000       1910 ns/op      352 B/op        5 allocs/op
BenchmarkVmihailencoMsgpackUnmarshal-8      1000000       2154 ns/op      352 B/op       13 allocs/op
BenchmarkJsonMarshal-8                       500000       3714 ns/op     1232 B/op       10 allocs/op
BenchmarkJsonUnmarshal-8                     500000       4125 ns/op      416 B/op        7 allocs/op
BenchmarkEasyJsonMarshal-8                  1000000       1920 ns/op      784 B/op        5 allocs/op
BenchmarkEasyJsonUnmarshal-8                1000000       1659 ns/op      160 B/op        4 allocs/op
BenchmarkBsonMarshal-8                      1000000       1886 ns/op      392 B/op       10 allocs/op
BenchmarkBsonUnmarshal-8                     500000       2487 ns/op      248 B/op       21 allocs/op
BenchmarkGobMarshal-8                       1000000       1282 ns/op       48 B/op        2 allocs/op
BenchmarkGobUnmarshal-8                     1000000       1306 ns/op      112 B/op        3 allocs/op
BenchmarkXdrMarshal-8                        500000       2740 ns/op      455 B/op       20 allocs/op
BenchmarkXdrUnmarshal-8                     1000000       2041 ns/op      239 B/op       11 allocs/op
BenchmarkUgorjiCodecMsgpackMarshal-8         500000       3316 ns/op     2753 B/op        8 allocs/op
BenchmarkUgorjiCodecMsgpackUnmarshal-8       500000       3289 ns/op     3008 B/op        6 allocs/op
BenchmarkUgorjiCodecBincMarshal-8            500000       3237 ns/op     2785 B/op        8 allocs/op
BenchmarkUgorjiCodecBincUnmarshal-8          500000       3631 ns/op     3168 B/op        9 allocs/op
BenchmarkSerealMarshal-8                     300000       4453 ns/op      912 B/op       21 allocs/op
BenchmarkSerealUnmarshal-8                   500000       3889 ns/op     1008 B/op       34 allocs/op
BenchmarkBinaryMarshal-8                    1000000       1841 ns/op      256 B/op       16 allocs/op
BenchmarkBinaryUnmarshal-8                  1000000       1945 ns/op      336 B/op       22 allocs/op
BenchmarkFlatbuffersMarshal-8               3000000        411 ns/op        0 B/op        0 allocs/op
BenchmarkFlatBuffersUnmarshal-8             5000000        354 ns/op      112 B/op        3 allocs/op
BenchmarkCapNProtoMarshal-8                 2000000        578 ns/op       56 B/op        2 allocs/op
BenchmarkCapNProtoUnmarshal-8               3000000        515 ns/op      200 B/op        6 allocs/op
BenchmarkCapNProto2Marshal-8                1000000       1427 ns/op      244 B/op        3 allocs/op
BenchmarkCapNProto2Unmarshal-8              1000000       1325 ns/op      320 B/op        6 allocs/op
BenchmarkHproseMarshal-8                    1000000       1294 ns/op      479 B/op        8 allocs/op
BenchmarkHproseUnmarshal-8                  1000000       1715 ns/op      320 B/op       10 allocs/op
BenchmarkProtobufMarshal-8                  1000000       1554 ns/op      200 B/op        7 allocs/op
BenchmarkProtobufUnmarshal-8                1000000       1055 ns/op      192 B/op       10 allocs/op
BenchmarkGoprotobufMarshal-8                2000000        746 ns/op      312 B/op        4 allocs/op
BenchmarkGoprotobufUnmarshal-8              2000000        978 ns/op      432 B/op        9 allocs/op
BenchmarkGogoprotobufMarshal-8             10000000        211 ns/op       64 B/op        1 allocs/op
BenchmarkGogoprotobufUnmarshal-8            5000000        289 ns/op       96 B/op        3 allocs/op
BenchmarkColferMarshal-8                   10000000        178 ns/op       64 B/op        1 allocs/op
BenchmarkColferUnmarshal-8                 10000000        235 ns/op      112 B/op        3 allocs/op
BenchmarkGencodeMarshal-8                   5000000        212 ns/op       80 B/op        2 allocs/op
BenchmarkGencodeUnmarshal-8                 5000000        273 ns/op      112 B/op        3 allocs/op
BenchmarkGencodeUnsafeMarshal-8            10000000        135 ns/op       48 B/op        1 allocs/op
BenchmarkGencodeUnsafeUnmarshal-8          10000000        198 ns/op       96 B/op        3 allocs/op

Issues

The benchmarks can also be run with validation enabled.

VALIDATE=1 go test -bench='.*' ./

Unfortunately, several of the serializers exhibit issues:

1. (minor) BSON drops sub-microsecond precision from time.Time.
2. (minor) Ugorji Binc Codec drops the timezone name (eg. "EST" -> "-0500") from time.Time.

--- FAIL: BenchmarkBsonUnmarshal-8
    serialization_benchmarks_test.go:115: unmarshaled object differed:
        &{20b999e3621bd773 2016-01-19 14:05:02.469416459 -0800 PST f017c8e9de 4 true 0.20887343719329818}
        &{20b999e3621bd773 2016-01-19 14:05:02.469 -0800 PST f017c8e9de 4 true 0.20887343719329818}
--- FAIL: BenchmarkUgorjiCodecBincUnmarshal-8
    serialization_benchmarks_test.go:115: unmarshaled object differed:
        &{20a1757ced6b488e 2016-01-19 14:05:15.69474534 -0800 PST 71f3bf4233 0 false 0.8712180830484527}
        &{20a1757ced6b488e 2016-01-19 14:05:15.69474534 -0800 -0800 71f3bf4233 0 false 0.8712180830484527}

All other fields are correct however.

Additionally, while not a correctness issue, FlatBuffers, ProtoBuffers and Cap'N'Proto do not support time types directly. In the benchmarks an int64 value is used to hold a UnixNano timestamp.