![]() |
![]() ![]()
Experimental Physics and
| ||||||||||||||
|
On Tue, 8 Jul 2025 at 06:29, lvhh at sari.ac.cn <lvhh at sari.ac.cn> wrote:
That's 4kByte per update - or do you mean 1024 * 4 doubles = 32kByte? Assuming 32kByte @ 50Hz that would be ~1.6MByte/s, which comfortably fits on a 100MBit connection. Network bandwidth is not an issue.
Assuming your AA is running on a server-type machine with enough memory and a fast disk, I would not expect an issue. But you should definitely test this: Run an IOC where a thread creates fake data in a waveform record of the required size. Make the fake data a pattern that allows you to easily detect a missing update in the archived data (like a counter). Let it run and archive for a day, then increase the rate, until you see updates missing. As long as your production setup stays well below that tested limit, you're good. Cheers, ~Ralph
| ||||||||||||||
ANJ, 08 Jul 2025 |
![]() · Download · Search · IRMIS · Talk · Documents · Links · Licensing · |