Skip to content

Sight_FrameServer is the delivery method used by SIGHT for remote visualization. SIGHT is an exploratory visualization tool for large scale datasets supporting manycore and multicore advanced shading, remote and interactive scientific visualization, parallel I/O and large scale displays. SIGHT is currently deployed in the OLCF systems to support…

License

Notifications You must be signed in to change notification settings

benjha/Sight_FrameServer

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

*External dependencies

- TurboJPEG Library (CPU based compression)
- LibPNG12 (a different version should work)
- NvPipe (H264 GPU based compression https://github.com/NVIDIA/NvPipe)


I. Compilation

$~/Sight_FrameServer/cd Release
$~/Sight_FrameServer/Release/make


Note: all auxiliary files for compilation are available in Release directory. If TurboJPEG Library and/or LibPNG12 are not installed in default locations edit makefile file to add the new paths.

NvPipe and Broadway.js Notes:
-Client uses Broadway.js for H264 decoding, thus only H264 encoding is supported using NvPipe.
-Enable NV_ENC_H264_PROFILE_BASELINE_GUID in NvPipe.
-Broadway.js uses webworkers for decoding. Google Chrome does not support webworker calls from file://, thus Mozilla
Firefox is recommended.


II. Run

1. Run the Server:

 ./Sight_FrameServer 

Sight_FrameServer will listen to port 9002. Make sure this port is open.

2. Run the Client

Open Sight_Client directory and double click on client.html file, then make click on "Stream" button. Sight_FrameServer will generate a 
random color image. 

Controls:

-Press keys '1' and '2' to swtich between rendering modes.
-Sight_FrameServer will output mouse callbacks. 
-Clicking on "Capture" button to save a snapshot, they will be saved in Sight_FrameServer directory.


III. Running Sight_FrameServer remotely in a Cluster

1. Server Configuration

Deploy Sight_FrameServer in the remote system.

Open Sight_FrameServer/header/cBroadCastServer.h file, comment FULLHD macro (code line 20) and uncomment REMOTE macro (code line 19). Follow
compilation instructions. 

2. Client Configuration

In your local system open Sight_FrameServer/Sight_Client/websocketConnection.js, set  jpegCompression variable (code line 18) to true. Save the file.

3. Once configuration is done, you will need two terminals.

In terminal 1:

- Connect to your remote server (and launch and interactive job).
- Check which node  is running  the interactive job (this will be used for  tunneling):
- Run Sight_FrameServer as explained in Section II, make sure 9002 port is open.

In Terminal 2  open a tunneling connection with the node is running the interactive job using port 9002:

$ ssh [email protected] -L 9002:myNodeName:9002

In your local system run the client as explained in Section II.

IV. Running Sight_FrameServer remotely in a server

Follow steps 1 and 2 of Section III.

3. Once configuration is done, you will need two terminals.

In terminal 1:

- Connect to your remote server.
- Run Sight_FrameServer as explained in Section II, make sure 9002 port is open.

In Terminal 2  open a tunneling connection with the server using port 9002:

$ ssh [email protected] -L 9002:myserver.com:9002

In your local system run the client as explained in Section II.


NOTICE:

If you use this code  in a publication, a link to or citation of this repository would be appreciated. The BibTex entry is

@misc{SIGHT_FrameServer,
  auhor = {Benjam\'in Hern\'andez},
  title = {SIGHT Frameserver: an approach for remote visualization.},
  year  = {2018},
  url   = {https://github.com/benjha/Sight_FrameServer},
  note	= {\url{https://github.com/benjha/Sight_FrameServer}}
}

About

Sight_FrameServer is the delivery method used by SIGHT for remote visualization. SIGHT is an exploratory visualization tool for large scale datasets supporting manycore and multicore advanced shading, remote and interactive scientific visualization, parallel I/O and large scale displays. SIGHT is currently deployed in the OLCF systems to support…

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published