Decode multi-stream with GTX 1080 using DeepStream SDK

Hi guys,
I have some question about multi-streaming decode with dGPU.

I want to run DeepStream SDK 4.0 on PC and to doing multi-stream decode :
I follow medium post about Use NVIDIA DeepStream to Accelerate H.264 Video Stream Decoding, and my system has such specifications:

GPU : GTX 1080
OS : ubuntu 16.04
TensorRT : 5.x
Cuda : 10.1
, …

when I run sudo apt-get install ./deepstream-4.0_4.0-1_amd64.deb, I get this error:

Reading package lists… Done
Building dependency tree
Reading state information… Done
Note, selecting ‘deepstream-4.0’ instead of ‘./deepstream-4.0_4.0-1_amd64.deb’
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:

The following packages have unmet dependencies:
deepstream-4.0 : Depends: libcairo2 (>= 1.15.10) but 1.14.6-1 is to be installed
Depends: libglib2.0-0 (>= 2.56.4) but 2.48.2-0ubuntu4.6 is to be installed
Depends: libgstreamer1.0-0 (>= 1.14.1) but 1.8.3-1~ubuntu0.1 is to be installed
Depends: libgstreamer1.0-dev (>= 1.14.1) but 1.8.3-1~ubuntu0.1 is to be installed
Depends: libgstreamer-plugins-base1.0-0 (>= 1.14.1) but 1.8.3-1ubuntu0.3 is to be installed
Depends: libgstreamer-plugins-base1.0-dev (>= 1.14.1) but 1.8.3-1ubuntu0.3 is to be installed
Depends: libnvinfer5 (>= 5.1.2) but it is not installable
Depends: libnvinfer-dev (>= 5.1.2) but it is not installable
Depends: libpangocairo-1.0-0 (>= 1.40.14) but 1.38.1-1 is to be installed
E: Unable to correct problems, you have held broken packages.

Q1 - What do i do for solving this problem?
Q2 - GTX 1080 has special hardware for decoding and encoding or use cuda? If it has special hardware for this task, why use GPU RAM? Also use CPU 11-12%, why use CPU? is’t done all on GPU?

Q1 - What do i do for solving this problem?

–> deepstream version 4.0 or above use ubuntu 18.04, you need to upgrade the os, or you can use docker to run.

Q2 - GTX 1080 has special hardware for decoding and encoding or use cuda? If it has special hardware for this task, why use GPU RAM? Also use CPU 11-12%, why use CPU? is’t done all on GPU?

–> GTX 1080 have dedicated hardware for decoding, please
see this, https://developer.nvidia.com/video-encode-decode-gpu-support-matrix#Decoder, search 1080
the pipeline you used include code which needs run on CPU and GPU. and please noted GPU and video engine share same device memory.

Thanks a lot.