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 GitHub Actions worker running Ubuntu:

Benchmark		Average time (ms)
Benchmark		Chr.Avro	Confluent.Apache.Avro
array[double]	deserialization 100,000 iterations	20.214	n/a
array[double]	serialization 100,000 iterations	41.771	n/a
array[double] (generic)	deserialization 100,000 iterations	n/a	38.513
array[double] (generic)	serialization 100,000 iterations	n/a	62.156
array[double] (specific)	deserialization 100,000 iterations	n/a	49.035
array[double] (specific)	serialization 100,000 iterations	n/a	62.660
boolean	deserialization 1,000,000 iterations	2.311	34.814
boolean	serialization 1,000,000 iterations	6.902	67.869
double	deserialization 1,000,000 iterations	30.924	58.412
double	serialization 1,000,000 iterations	42.508	110.496
fixed[12]	deserialization 1,000,000 iterations	19.187	n/a
fixed[12]	serialization 1,000,000 iterations	25.234	n/a
fixed[12] (generic)	deserialization 1,000,000 iterations	n/a	73.483
fixed[12] (generic)	serialization 1,000,000 iterations	n/a	74.433
fixed[12] (specific)	deserialization 1,000,000 iterations	n/a	88.462
fixed[12] (specific)	serialization 1,000,000 iterations	n/a	34.325
fixed[2048]	deserialization 100,000 iterations	59.988	n/a
fixed[2048]	serialization 100,000 iterations	63.224	n/a
fixed[2048] (generic)	deserialization 100,000 iterations	n/a	63.525
fixed[2048] (generic)	serialization 100,000 iterations	n/a	63.760
fixed[2048] (specific)	deserialization 100,000 iterations	n/a	68.915
fixed[2048] (specific)	serialization 100,000 iterations	n/a	63.502
float	deserialization 1,000,000 iterations	22.998	54.323
float	serialization 1,000,000 iterations	32.756	78.465
int	deserialization 1,000,000 iterations	9.658	46.015
int	serialization 1,000,000 iterations	18.911	64.618
long	deserialization 1,000,000 iterations	12.969	59.512
long	serialization 1,000,000 iterations	34.380	76.131
map[double]	deserialization 100,000 iterations	37.479	n/a
map[double]	serialization 100,000 iterations	41.359	n/a
map[double] (generic)	deserialization 100,000 iterations	n/a	51.784
map[double] (generic)	serialization 100,000 iterations	n/a	64.784
map[double] (specific)	deserialization 100,000 iterations	n/a	58.827
map[double] (specific)	serialization 100,000 iterations	n/a	83.686
record	deserialization 100,000 iterations	97.870	n/a
record	serialization 100,000 iterations	87.352	n/a
record (generic)	deserialization 100,000 iterations	n/a	60.221
record (generic)	serialization 100,000 iterations	n/a	39.633
record (specific)	deserialization 100,000 iterations	n/a	33.984
record (specific)	serialization 100,000 iterations	n/a	33.804
string	deserialization 1,000,000 iterations	74.759	99.088
string	serialization 1,000,000 iterations	79.096	125.743

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

Chr.Avro

Performance and benchmarks

Guides

Technical documentation

CLI reference

API reference