Sorry for my english because it’s not the best but i will try to explain me.
I have several clients with more than 100 GB of data, but when I try to recover all the data, the download goes to 3Mbs (I download it as ZIP), each restore takes about 24 hours or more, when I look at the server utilization it is at 20% the CPU and RAM, I have the entire network 1000, switch router, cat6, everything, the hard drive is WD Barracuda and it is between 150-200 MBs, is there a way to restore faster, the network supports it. The Server OS it a Windows Server 2019
Do you have any kind of virus scanner running that could be slowing things down scanning that amount of data? Is the client running low on CPU or RAM resources? 20% CPU and RAM utilization on the server does not sound bad. Do you have any I/O bottlenecks on either the server or the client? I’m not sure how to display disk I/O throughput on a Windows computer because I am a Linux person.
You could do some testing of smaller restores - maybe 1Gb or 10Gb - to see what your speeds are. Does your download always drop to 3Mbs regardless of the size of the download? If your speed is only dropping for very large downloads, that would be a clue. What that clue would indicate, I don’t know. But it’s at least a clue that you could research. You could also run some tests of large data transfers NOT using UrBackup, check your speeds for those, to look for clues if what you have is a general network problem or something more specific to UrBackup and what it’s doing (ZIP-ing and UNZIP-ing, etc.)
You could try sequentially rebooting each involved computer and then the router to see if that makes any difference. Sometimes computers and routers just start misbehaving for reasons that you may never figure out.
Hello, I have tried, everything you have told me, I have tried on all my servers (I have 3 servers), everything gives the same result, both small and large files, if I do it by tcp, remote or something similar, I pass the data to more than between 40 to 60MBs approx, I think I have the I/O bottlenecks because I have more than 700ms, when I have seen that the acceptable is less than 10ms, (I have 2 the same (Xeon 2420, 48 Gb Ram Ddr3 and 1 different ( Xeon 2620 v3 , 48 Gb Ram Ddr4 ) )