Upgraded to 2.5.1 - slow response with AI results?
-
Upgraded to 2.5.1 today. Some small hiccups but got those sorted and everything seems to operational. While watching the logs I noticed that AI response is (to me) quite high (300ms++) based on other feedback of other users. One item I do notice is that cuDNN: 8.5 is being reported even though I did a fresh install of the cudnn batch script which should be 8.9 (both 8.5 and 8.9 directories are there under nvidia). Is this a valid outcome, or did I miss/screw something up? :-) System Info tab: ========================================= Server version: 2.5.1 System: Windows Operating System: Windows (Microsoft Windows 10.0.19045) CPUs: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz (Intel) 2 CPUs x 2 cores. 2 logical processors (x64) GPU (Primary): Quadro P2000 (5 GiB) (NVIDIA) Driver: 522.06, CUDA: 11.8 (up to: 11.8), Compute: 6.1, cuDNN: 8.5 System RAM: 16 GiB Platform: Windows BuildConfig: Release Execution Env: Native Runtime Env: Production .NET framework: .NET 7.0.5 Default Python: Video adapter info: Red Hat QXL controller: Driver Version 10.0.0.18000 Video Processor QEMU QXL Microsoft Remote Display Adapter: Driver Version 10.0.19041.3636 Video Processor NVIDIA Quadro P2000: Driver Version 31.0.15.2206 Video Processor Quadro P2000 System GPU info: GPU 3D Usage 1% GPU RAM Usage 1.1 GiB Global Environment variables: CPAI_APPROOTPATH = CPAI_PORT = 32168 =============================== Clip from server log: =============================== 19:49:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'detect' (...0f689b) took 367ms 11:19:49:Response received (#reqid 84201b96-8a0e-41dd-b589-b48d350f689b for command detect): Found truck, car, truck 11:19:49:Client request 'detect' in queue 'objectdetection_queue' (...17fa1a) 11:19:49:Request 'detect' dequeued from 'objectdetection_queue' (...17fa1a) 11:19:49:Object Detection (YOLOv5 6.2): Retrieved objectdetection_queue command 'detect' 11:19:49:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'detect' (...17fa1a) took 267ms 11:19:49:Response received (#reqid 16651331-35f9-4141-a57d-36468b17fa1a for command detect): No objects found 11:20:16:Client request 'detect' in queue 'objectdetection_queue' (...b61f10) 11:20:16:Request 'detect' dequeued from 'objectdetection_queue
-
Upgraded to 2.5.1 today. Some small hiccups but got those sorted and everything seems to operational. While watching the logs I noticed that AI response is (to me) quite high (300ms++) based on other feedback of other users. One item I do notice is that cuDNN: 8.5 is being reported even though I did a fresh install of the cudnn batch script which should be 8.9 (both 8.5 and 8.9 directories are there under nvidia). Is this a valid outcome, or did I miss/screw something up? :-) System Info tab: ========================================= Server version: 2.5.1 System: Windows Operating System: Windows (Microsoft Windows 10.0.19045) CPUs: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz (Intel) 2 CPUs x 2 cores. 2 logical processors (x64) GPU (Primary): Quadro P2000 (5 GiB) (NVIDIA) Driver: 522.06, CUDA: 11.8 (up to: 11.8), Compute: 6.1, cuDNN: 8.5 System RAM: 16 GiB Platform: Windows BuildConfig: Release Execution Env: Native Runtime Env: Production .NET framework: .NET 7.0.5 Default Python: Video adapter info: Red Hat QXL controller: Driver Version 10.0.0.18000 Video Processor QEMU QXL Microsoft Remote Display Adapter: Driver Version 10.0.19041.3636 Video Processor NVIDIA Quadro P2000: Driver Version 31.0.15.2206 Video Processor Quadro P2000 System GPU info: GPU 3D Usage 1% GPU RAM Usage 1.1 GiB Global Environment variables: CPAI_APPROOTPATH = CPAI_PORT = 32168 =============================== Clip from server log: =============================== 19:49:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'detect' (...0f689b) took 367ms 11:19:49:Response received (#reqid 84201b96-8a0e-41dd-b589-b48d350f689b for command detect): Found truck, car, truck 11:19:49:Client request 'detect' in queue 'objectdetection_queue' (...17fa1a) 11:19:49:Request 'detect' dequeued from 'objectdetection_queue' (...17fa1a) 11:19:49:Object Detection (YOLOv5 6.2): Retrieved objectdetection_queue command 'detect' 11:19:49:Object Detection (YOLOv5 6.2): Rec'd request for Object Detection (YOLOv5 6.2) command 'detect' (...17fa1a) took 267ms 11:19:49:Response received (#reqid 16651331-35f9-4141-a57d-36468b17fa1a for command detect): No objects found 11:20:16:Client request 'detect' in queue 'objectdetection_queue' (...b61f10) 11:20:16:Request 'detect' dequeued from 'objectdetection_queue