User Tools

Site Tools


faq

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
faq [2023/11/13 11:54]
nightfly
faq [2024/04/18 14:50] (current)
nightfly
Line 15: Line 15:
  
 **Q:** Can you give us some examples of the possible number of cameras per server?\\ **Q:** Can you give us some examples of the possible number of cameras per server?\\
-**A:** In real life, we already have NVRs running on used servers, 8-10 years old, with few hundred cameras on them.\\+**A:** In real life, we already have NVRs running on used servers, 8-10 years old, with few hundred cameras on them. Yep, thats about 0.4-0.5% per camera one core load, on CPU like Intel Xeon E5-2430L or about 0.6-0.7% core/per camera load on CPU like Intel Xeon x3440 from 2009.\\
  
 **Q:** How is the available storage space distributed among the video captured from the cameras?\\ **Q:** How is the available storage space distributed among the video captured from the cameras?\\
-**A:** Equally within each storage.\\+**A:** Equally within each storage. By default for video archive is reserved 90% of storage space.\\ 
 + 
 +**Q:** How can I determine the load on my storage? I mean, how can I determine if Disk IO or write speed is a bottleneck?\\ 
 +**A:** Well, for example, you can use the gstat command.
  
 **Q:** What regulates how long the video archive of each camera is stored?\\ **Q:** What regulates how long the video archive of each camera is stored?\\
Line 36: Line 39:
  
 **Q:** What camera recording modes or schedules are supported?\\ **Q:** What camera recording modes or schedules are supported?\\
-**A:** Only continuous recording. Our position is that if the camera is turned on, all data from it should be saved and available at any time when it is needed.\\+**A:** Only continuous recording. Our position is that if the camera is enabled, all data from it should be saved and available at any time when it is needed.\\
  
 **Q:** What about video object recognition and motion detection?\\ **Q:** What about video object recognition and motion detection?\\
-**A:** Currently, we are developing the implementation of both object recognition and separation of video parts with motion using neural networks. But it will work only for already exported video files. It is quite obvious that we are not ready to sacrifice the performance of the entire system to do this in real time.\\+**A:** Currently, we are developing the implementation of both object recognition and separation of video parts with motion using neural networks. But it will work only for already exported video files on demand. It is quite obvious that we are not ready to sacrifice all performance of the entire system to do this in real time.\\
  
  
faq.1699869246.txt.gz · Last modified: 2023/11/13 11:54 by nightfly