When using Wireguard and a VPN physically from a laptop — when I'm traveling — to virtually use my home desktop, am I using the processing cores/RAM in the laptop or in the (more powerful) desktop when running multicore processes?
I'm posting here because I got the Wireguard information from this subreddit. Also, I am an academic and a computational researcher, so I want to go to the best-of-the-best (you all) and also really learn about how VPNs work. Security is important, as I am a researcher. I am hoping for both the simple answer and the in-depth teaching answer, if you are willing to spend the time. I greatly appreciate it in advance.
Basically, I went all out (that I could afford) on my home desktop where I have 16 cores (Windows) and can ssh into two 96-core workstations and a 48-core workstation (all 3 Linux). So, I already have all of the power I need to run my work. What I don't have is a lightweight device that can utilize all those powerful units while I'm on-the-go. I don't want to really pay for more processing power if I don't need to. I'm hoping that I can essentially use my desktop virtually while physically away from it. I haven't yet purchased a laptop, but I'm hoping to budget the processing power when I do have to spend the money on the screen due to visual impairments; the laptop will be Windows.
My questions are essentially these:
(1) Can I use that VPN set-up to virtually use my 16 cores and 32 GB of RAM in my desktop while physically on, say, a 4-core and 8 GB laptop? None of my work is extremely graphically intensive, but the simulations are multicore and take some power.
(2) I'm hoping I can access the files on the desktop while on-the-go and if I change some file and save it while using the VPN set-up above while physically on the laptop, that it will also be changed and saved on the desktop when I return. Is this the case?
(3) I am a little confused about VPNs. My university requires that I VPN (Cisco) into their system prior to SSHing into the workstations that are connected to their system. Before SSHing, I am still on my desktop while logged on to their VPN. (I think it is just the internet usage, as I have permissions for journals that I don't without their VPN, but I'm definitely using my desktop's power when running code and am seeing and making changes to my desktop.) But, based on the other question that I read here regarding accessing the desktop from the laptop, it seems like the VPN set-up above allows one to essentially do what I do with the SSHing into the workstations without creating a shell, and instead actually be connected to and making changes on the virtual device. How does this work? Please teach me! (Feel free to skip this question if you don't have the time!)
[link] [comments]
from hacking: security in practice https://ift.tt/xdG7U8m
Comments
Post a Comment