Benchmarks

Policy enforcement overhead is benchmarked in model_b_test.go. Testing environment:

Intel(R) Core(TM) i7-6700HQ CPU @ 2.60GHz, 2601 Mhz, 4 Core(s), 8 Logical Processor(s)

Results from go test -bench=. -benchmem (op = one Enforce() call, ms = millisecond, KB = kilobytes):

Test senaryosu	Kural boyutu	Zaman ek yükü (ms/op)	Bellek ek yükü (KB)
ACL	2 kural (2 kullanıcı)	0.015493	5.649
RBAC	5 kural (2 kullanıcı, 1 rol)	0.021738	7.522
RBAC (küçük)	1100 kurallar (1000 kullanıcı, 100 roller)	0,164309	80,620
RBAC (orta)	11000 kurallar (10000 kullanıcı, 1000 roller)	2,258262	765,152
RBAC (büyük)	110000 kurallar (100000 kullanıcı, 10000 roller)	23,916776	7.606
Kaynak rolleri ile RBAC	6 kural (2 kullanıcı, 2 rol)	0,021146	7,906
Alanlar/kiracılar ile RBAC	6 kural (2 kullanıcı, 1 rol, 2 alan)	0,032696	10,755
ABAC	0 kural (0 kullanıcı)	0.007510	2.328
RESTful	5 kural (3 kullanıcı)	0.045398	91.774
Reddet-üstünlük	6 kural (2 kullanıcı, 1 rol)	0.023281	8.370
Öncelik	9 kural (2 kullanıcı, 2 rol)	0.016389	5.313

Policy enforcement overhead in Pycasbin is benchmarked in tests/benchmarks. Testing environment:

Intel(R) Xeon(R) CPU E5-2673 v4 @ 2.30GHz (Run by Github actions)
platform linux -- Python 3.11.4, pytest-7.0.1, pluggy-1.2.0

Results from executing casbin_benchmark (op = one enforce() call, ms = millisecond):

Policy enforcement overhead in Casbin CPP is benchmarked in tests/benchmarks using Google's benchmarking tool. Testing environment:

Intel(R) Core(TM) i5-6300HQ CPU @ 2.30GHz, 4 cores, 4 threads

Results from executing the casbin_benchmark target built in Release configuration (op = one enforce() call, ms = millisecond):

Policy enforcement overhead in Lua Casbin is benchmarked in bench.lua. Testing environment:

AMD Ryzen(TM) 5 4600H CPU @ 3.0GHz, 6 Cores, 12 Threads

Results from running luajit bench.lua (op = one enforce() call, ms = millisecond):

Benchmark izleme

The embedded page below displays Casbin's performance across commits.