Meanwhile the managers, (I only assume), got huge bonuses out of it.
Things like this happen when the person who signs the check is not the person who actually uses the software. Also, no matter the resulting clusterfuck, the project is always successful.
Then COVID hit and all of a sudden we all worked from home exactly the same way. It was a wow moment of how prepared we were comparing to everyone.
Yes, it is never as good as native, but it's not that bad if done properly.
Citrix/RDP farm should NEVER be used for cost savings. It is for data control. But bean counters think it saves costs... and user experience suffer.
Even even over a paltry 15mbit consumer grade cable internet uplink it is possible to play full screen YouTube videos in high def over plain vanilla RDP version 8, without even getting into any of the (often quite flaky and at this point highly redundant) Citrix enhancements.
Well.. for starters, it's fast only if your datacenter where the RDP/Citrix server is close to the terminals. If ping time is 100ms round trip, then yeah, it's gonna be slow, and it's going to begin to resemble dial-up.
> ... or criminally oversubscribed
Yes! and people got bonuses out of it!
I was apprehensive at first but actually it's pretty great. I can just turn off my computer at the office and turn it back on at home and get back the same session. I can move to any stations in the building without issues.
I don't need to carry around a laptop so I can go to the office with just my keycard and my wallet. I can buy whatever device I want for myself and use it for work too without actually mixing personal and work related computing at all.
I've heard it costs more than buying new laptops to everyone every year.
If only "costs more" if you look at a pure dollar cost of the VDI infra vs the pure cost of a laptop - yeah, it might be more than a couple hundred $$ for for the infra to manage it every month vs a new $2400 laptop every year
But updates, preferences, personalization, setup time, etc is all handled - on a schedule, in a unified manner, and uniformly for everyone
First phase of a tech rollout gives overly generous hardware, because the end user needs to be won over. Then comes right-sizing because it makes economic sense, and we know things are too generous. Then end users get decoupled from bean counters and the provisioning gets done by a third party corporation, so feedback cycles are broken. At that point, the race to the bottom starts.
I've seen people declare: We provision to the satisfaction of the customer. Which means: If you don't do the effort to complain, all is well. The results is the best people leave, as they won't deal with that. The middlings complain, bothering their managers, so they get labelled troublemakers and are oushed out. The yes men stay and praise the situation. The 9to5ers stay and get paid no matter what, so who cares if msword's text appear multiple seconds after you typed it.
But the TCO of employee connectivity is not merely "the device"
It's everything else that goes into it
Want a standard developer environment? Create an appropriate VDI instance
Want a standard CAD environment? Create an appropriate VDI instance
Want a standard xyz environment? Create an appropriate VDI instance
Then, based on the user role (at the moment of login), spin-up and connect to the proper image
This also allows someone who does two jobs to connect to multiple VDIs based off work needs, and do the 'proper' work on each one
Sure some of this is about data governance ... but most is about enabling end users to be as productive as possible
Updates? Done offline in offhours just to the master image(s)
New version of $OS$ needs to be deployed/trialed? Build/update one image at a time, verify, then push out as the new Golden Master
That CAD program that needs 16 threads for someone to be productive? But if the cluster is oversubscribed you'll get maybe 30% of the performance you need.
Except ... it does "save costs"
Allowing BYOD (or eve just supplying a 'basic' device) and connecting only through Horizon (or whatever client of choice you prefer) is a huge savings over managing fleets of laptops, desktops, etc - all with different support cycles, specs, etc
All your management is focused on just one layer - the virtual desktop (and its supporting infrastructure)
Everything else is noise
You'll have a lot of thin-client/remote environments used for places where the primary metric is manageability. Think of how schools love Chromebooks, and places like call centres that simply want users in a bolted down kiosk mode. They don't care if the experience is miserable, but they like the idea that nobody's reading Lemmy.
It's going to be a huge hairball for the more advanced user market. Your engineering and graphics and big-data people are going to need special tools with likely complex licensing and setup hassles, mitigating the manageability benefits, and you know there will be a point where hiring a big cloud machine gets pricier than buying an actual workstation.
I wonder if some vendors will instead buy into a reduced-version of the concept for distribution though-- instead of ever getting a copy of the bits for AutoCAD or Photoshop, you just get a RDP login to a big machine running it, almost reinventing the X11 paradigm.
I had the unfortunate pleasure of working on remote windows at JPMC. It was a nightmare. Yes they got a huge immediate cash savings on hardware, but at the cost of worker productivity.
The response time from keypress to response was literally like I was in 1992 working with a dialup connection to a terminal server.
Oh, also the executives got their own laptops, so they didn't dogfood their own solution.