Skip to content

CARS API Performance#

Test Server Configuration#

The efficiency of recognition of vehicle or LP attributes depends on the parameters of the input image:

  • Image size;
  • Number of bits per color.

Input data:

  • Vehicle image resolution: 500х439 px;
  • LP image resolution: 117x33 px;
  • Format of the original image: jpeg.

The test server parameters are shown in Table 31.

Table 31. Test Server Options

№ Resource Values
1 CPU Model name: Intel(R) Xeon(R) Gold 6240R CPU @ 2.40GHz, Thread(s) per core: 2, Core(s) per socket: 24
2 CPU frequency 2.40GHz
3 RAM 251Gb
4 Memory 240GB INTEL SSDSC2KB24
5 Video memory 15109MiB
6 Operation system CentOS 8

CARS API performance measurements are presented for several architectures (Table 32).

Table 32. Tests

Name Description
CPU Running the CARS API on a server with a central processing unit (CPU) without support for Advanced Vector Extensions 2 (AVX2) instructions
AVX2 Running the CARS API on a server with a CPU with support for AVX2 instructions
GPU Running the CARS API on a server with GPU

The measured values are the average of at least 100 experiments.

All available CPU and GPU cores were used in the experiments to determine the attributes of vehicles and LPs, listed in the tables below.

Performance test results#

The overall CARS API benchmark results are shown in Table 33. All values presented in milliseconds.

Table 33. Results

Classifier CPU AVX2 GPU
Car_brand_model_v1 119 35 37
Vehicle_color 33 15 13
Vehicle_emergency_type 30 16 16
Vehicle_type 121 33 35
Public_transport_type 142 35 36
Special_transport_type 145 35 37
Grz_all_countires 86 26 32