Multi-site installation & Push vs Pull
Posted: Wed May 07, 2014 7:30 am
Hi all,
I am a developer and am considering deploying ZM for a multi-site installation. We have a few different locations and will be needing multiple cameras at each. Had some questions before getting into the code:
1) In a ZM setup can IP cameras 'push' video to the server, or does the server always need to 'pull' from the cameras? If each camera can push video it would make it simpler so I can have one beefy server in our datacenter, and then just configure each camera individually to push data to it so I do not have to deal with any routing issues within each location.
2) In a multi-site deployment, do people usually have one server pull data from each camera at each location, or is there some notion of pre-processing where you have a server at each location doing processing to reduce the amount of data needing to be sent over the WAN to the master server that is doing archiving/consolidation for each site and is running the web interface for end users to access?
3) How does data ultimately get stored and archived? Is it frame by frame, or is it in videos for a given amount of time? I am wondering if similar views (e.g. when there is no change in activity) can get compressed or deduplicated out to reduce long-term storage size on large systems, or if every frame still gets stored.
Any input or feedback on multi-site deployments would be greatly appreciated. We are an engineering based company, so if we end up using ZM will likely contribute code back. Just trying to figure out what we will need to build on our own and what is already there.
Thanks,
Jordan
I am a developer and am considering deploying ZM for a multi-site installation. We have a few different locations and will be needing multiple cameras at each. Had some questions before getting into the code:
1) In a ZM setup can IP cameras 'push' video to the server, or does the server always need to 'pull' from the cameras? If each camera can push video it would make it simpler so I can have one beefy server in our datacenter, and then just configure each camera individually to push data to it so I do not have to deal with any routing issues within each location.
2) In a multi-site deployment, do people usually have one server pull data from each camera at each location, or is there some notion of pre-processing where you have a server at each location doing processing to reduce the amount of data needing to be sent over the WAN to the master server that is doing archiving/consolidation for each site and is running the web interface for end users to access?
3) How does data ultimately get stored and archived? Is it frame by frame, or is it in videos for a given amount of time? I am wondering if similar views (e.g. when there is no change in activity) can get compressed or deduplicated out to reduce long-term storage size on large systems, or if every frame still gets stored.
Any input or feedback on multi-site deployments would be greatly appreciated. We are an engineering based company, so if we end up using ZM will likely contribute code back. Just trying to figure out what we will need to build on our own and what is already there.
Thanks,
Jordan