Performance and benchmarks

To ensure that changes to Chr.Avro don’t introduce major performance regressions (and to highlight areas where performance could be improved), we maintain a set of rudimentary benchmarks to test serialization throughput.

Each benchmark runs five times, and we take the average time of those runs. These results were collected on a 2017 MacBook Pro (3.1 GHz Intel Core i5):

Benchmark		Average time (ms)
Benchmark		Chr.Avro	Confluent.Apache.Avro
array[double]	deserialization 1,000,000 iterations	217.578	n/a
array[double]	serialization 1,000,000 iterations	290.528	n/a
array[double] (generic)	deserialization 1,000,000 iterations	n/a	357.041
array[double] (generic)	serialization 1,000,000 iterations	n/a	552.456
array[double] (specific)	deserialization 1,000,000 iterations	n/a	427.759
array[double] (specific)	serialization 1,000,000 iterations	n/a	530.058
boolean	deserialization 10,000,000 iterations	44.248	218.248
boolean	serialization 10,000,000 iterations	91.741	354.362
double	deserialization 10,000,000 iterations	232.562	407.173
double	serialization 10,000,000 iterations	358.408	758.225
fixed[12]	deserialization 10,000,000 iterations	186.920	n/a
fixed[12]	serialization 10,000,000 iterations	176.214	n/a
fixed[12] (generic)	deserialization 10,000,000 iterations	n/a	553.247
fixed[12] (generic)	serialization 10,000,000 iterations	n/a	491.722
fixed[12] (specific)	deserialization 10,000,000 iterations	n/a	881.176
fixed[12] (specific)	serialization 10,000,000 iterations	n/a	261.225
fixed[2048]	deserialization 1,000,000 iterations	556.955	n/a
fixed[2048]	serialization 1,000,000 iterations	813.730	n/a
fixed[2048] (generic)	deserialization 1,000,000 iterations	n/a	651.146
fixed[2048] (generic)	serialization 1,000,000 iterations	n/a	1440.562
fixed[2048] (specific)	deserialization 1,000,000 iterations	n/a	716.324
fixed[2048] (specific)	serialization 1,000,000 iterations	n/a	866.015
float	deserialization 10,000,000 iterations	186.185	392.221
float	serialization 10,000,000 iterations	266.083	534.501
int	deserialization 10,000,000 iterations	143.195	356.727
int	serialization 10,000,000 iterations	212.832	488.547
long	deserialization 10,000,000 iterations	228.732	497.893
long	serialization 10,000,000 iterations	379.233	617.385
map[double]	deserialization 1,000,000 iterations	416.796	n/a
map[double]	serialization 1,000,000 iterations	359.202	n/a
map[double] (generic)	deserialization 1,000,000 iterations	n/a	534.978
map[double] (generic)	serialization 1,000,000 iterations	n/a	570.418
map[double] (specific)	deserialization 1,000,000 iterations	n/a	605.435
map[double] (specific)	serialization 1,000,000 iterations	n/a	608.235
record	deserialization 1,000,000 iterations	647.625	n/a
record	serialization 1,000,000 iterations	1311.111	n/a
record (generic)	deserialization 1,000,000 iterations	n/a	590.830
record (generic)	serialization 1,000,000 iterations	n/a	427.407
record (specific)	deserialization 1,000,000 iterations	n/a	318.839
record (specific)	serialization 1,000,000 iterations	n/a	283.461
string	deserialization 10,000,000 iterations	616.576	749.068
string	serialization 10,000,000 iterations	835.419	1213.418

In the future, we’re hoping to expand the benchmarks to test more complex situations as well as non-.NET libraries.

Performance and benchmarks

Guides

Technical documentation

CLI reference

API reference