Slashdot Mirror
Would You Move to Windows Thin Clients?
"Most users will be running basic MS Office apps, Groupwise for e-mail, and accessing some Oracle databases. A consultant hired for preliminary recommendations is saying that we should run Windows XP on the thin client boxes, not even the embedded version but the full one. Additionally, some of our users have more powerful applications like AutoCAD and ArcMap. We have already determined that those users will not be moving to the thin client machines.
Our department has spoken with a Citrix support/sales person who claims you can support up to 1000 clients on a single Citrix server. That seems so far from what I have generally read that I have a hard time buying it. Can anyone corroborate that claim? Again, most users will be using Office, Groupwise, and accessing Oracle DBs.
Does anyone have any experience with a workplace making this sort of migration? I would love to find a way to make it work, but from the research I have done so far, it doesn't look like we are going to get any cost-savings (unless they miraculously decide to go with Linux)."
>The whole setup is more centralized, making it easier to address Windows patches and virus updates.
I've always heard Windows thin clients described as a moving around of costs, not a cost-saver. The money you would have spent in desktops gets spent in servers+Citrix licenses. You don't have help desk people going out into the field to troubleshoot as much, but you have people spending time troubleshooting the servers. The real payoff is with the centralization.
Okay, no I'm not.
... or at least, don't change it much. Maybe re-think your use of networking and other such things, but still within the framework of your existing OS setup.
Why change?
What do you need to 'fix' in the existing system?
Continue to use per-desk PC's, albeit cheaper/better/easier to administer, and continue to utilize 'upgrade' licensing schemes - the cheapest you can possibly get from Microsoft, is, probably, very cheap - to maintain the OS upgrade/fix end of things, and you can probably do this all very cheap.
"Thin-client" is just another word for extremely small/light/cheap PC-in-a-box. So maybe the simpler option to 'considering remote client solutions' is simply, put it all into better hardware (new CPU/more RAM/*standard* video for all systems/etc.) and don't change your existing software standards
If you maintain your existing stance with regard to how well your business problems are solved by your software systems, and give it a 'boost' with periodical hardware upgrades for key areas/servers every now and then, there is little reason to drastically change everything at all.
When the PC is akin to the size/heft of a block of printer paper, 'administration' becomes relatively trivial - particularly in an organization of 2,000 people or so... if you've got your software worked out.
Just get smaller, lighter, cheaper PC's, and refresh them every now and then. Convert as much as possible to cheap laptops and monitors, even.
They are out there.
; -- the corruption of government starts with its secrets. a truly free people keep no secrets. --
First, why do you want to do this in the first place? You can get many of the benefits of "Thin Clients" by using a networked filesystem to store applications and configuration data, but still run the applications on the client PC. Also, you'll probably need less servers for this approach, since they'll just be networked disk servers, not application servers.
In my experience, the big benefit of having a Citrix server is being able to run Windows applications from clients that are not x86-based. For example, if the client is running Solaris on SPARC, but needs access to a Windows-only application, then that client can just access it through Citrix. Then any slowness from the network or the (loaded) server is somewhat acceptable, seeing as how you couldn't otherwise run the application. But that doesn't sound like the case at all in your scenario.
Also, you might want to see what VMWare has to offer; they are also in the x86 Server virtualization market.
pb Reply or e-mail; don't vaguely moderate.
...at $WORK I have a Linux box that handles ~25 full screen X sessions all day long, and hovers around .10 load average. The box is an old dual PIII 500MHz with 2GB RAM, 36GB RAID 0+1 volume. All of the client machines are Windows XP desktops using VNC viewer locally to attach to Xvnc sessions on the Linux box (which is kicked off in xinetd on-demand).
The server itself was considered too slow to run Windows. It had been retired from the Windows domain, and the bean counters have written it off the books. Yet there it is, serving a vital business function to my users that is about to scale up to another ~30 users in the next couple of weeks.
Every bit of software involved in making this happen was free. The hardware was, effectively, free. And I'm already handling more users than the newfangled expensive Windows 2000 Terminal Server that is parked in the same rack.
This move was made to stave off the sudden onslaught of requests for a second machine (linux) at every desk. The corporate standard desktop has been and will be Windows for some time to come. But setting up a Linux box with lots of RAM, fast disk and Xvnc has already saved us over $45,000 that would have otherwise been spent on dedicated Linux machines at every desk.
What do the average windows admin do? problems arise, reghost the image from the second partition, apply all new patches and youre set to go.
What do korean Cyber Cafe owners do? Theres some korean software better than PCSecure that allows access to all programs but renders the entire system read-only except the directories where you need to write. And then the admin remotely deletes those directories.
So admining over 70 systems, I can really understand your need to be in full control of all those systems without hiring a team of highschoolers. Its much worse if various clients need different software, many expect full control of their system and noone tolerates a slow boot. UNIX fixes these problems so beautifully.
I did some research years ago on remote-booting Windows98 maybe on gigabit ethernet and powerful servers. Theres a linux howto on how windows can be copied on a freshly made partition after a net boot, then booted. So it all really depends on each situation.
You mentioned XP. To decrease maintenance on hundereds of machines, use something more reliable and tested, like Windows98 or Win2k.
We use terminal services on all the 70 workstations with about 20 working at any time, and it works perfectly. To be fair, I'll tell you the server is an IBM eSeries with 2x PentiumIIIs and 2gigs of ram and 10k cheetah disks. On Pentium1 clients, the apps run much better on TS than native.
For any network based solution make sure your pipes are fat. Get good switches, tune them and do ethereal to test the switch-based traffic, make sure all clients have the good 3com NICS, should be all PCI 100 speed at least, and if you can help it, start with gigabit ethernet.
Oh yeah, since you're asking slashdot for help, if you do succeed in a new solution, you're obliged to submit a howto.
Cheers.
"Give orange me give eat orange me eat orange give me eat orange give me you." -Nim Chimpsky
I've setup 2 networks using WTS/Citrix. The first was back in the days of WinFrame (NT 3.51) at a polytechnic and the second is where I work now.
:-)
We have all 80 staff around the country (currently) connecting to 1 Terminal server for Office 97, Outlook, membership system. We do have a second TS that runs PageMaker and PhotoShop. And a third waiting to be rebuilt to create a wee bit of redundancy
Although our desktops are Win 2k Pro (cut right back and run the ICA and RDP clients) we have only 2 applications - one needs a local modem and the other is DOS based - that run on the local machines (apart from the laptops). We are starting to look at replacing the desktops with true thin clients (we bought the fat clients in case we had to go back to the 'old' way of doing things with the apps locally). The people with Palm devices can synchronise with Outlook/Exchange through the Citrix session. We have found that the people using laptops often prefer to dialup and connect to a terminal server rather than work off the local apps.
It means that all our data is on one server, not spread about the country. Everyone runs the same versions. We can shadow the user sessions. People can connect to exactly the same set of apps and have access to the same files, in the same locations from another office, from home, from the laptop in a motel room.
The benefits of thin client/terminal server whether it be Windows or Unix based is the only (well ok, a bit of exaggeration there) way to go.
Other people have already commented on the number of users per server. Short form: buy decent servers and assume 50:1 for now. Use Citrix, not Windows Terminal Services.
Look at costs and ROI and decide on a minimum acceptable ratio before you start. Is 10:1 worth it? How about 20:1? 50:1? 100:1?
The important thing is that you TEST before you deploy across the company. Find a few people IN DIFFERENT DEPARTMENTS who are willing to help (or be coerced) with your testing. Different departments are important because you want all of the pieces tested. For the salespeople, Powerpoint matters; for the bean counters, Excel. Buy one decent server. Worst case is that you get to replace the oldest machine in the shop a year sooner than planned. Convince Citrix to give you a trial license for a few months. Tell them that there'll be plenty of other purchases if the trial goes well.
Dedicate at least one full-time person to setting up the server. Remember- you're on the clock with the trial license. Get a client set up ASAP. Do what work you can through the client -- it's more testing. Deploy 5-10 clients for a week (few enough that you can visit all of them in-person in a short amount of time if there are problems) while you iron out replication and performance issues. If all's good, add another 10 each week in a controlled test roll-out until things bog down. If things are looking good, pull in any documentation or training people NOW, before the test users get too comfortable with the system.
Once things have bogged down, look at your target number. If you beat it, great! Run with it. If you didn't, how close were you? If you need 100:1 and you bogged down at 20:1, it's time to give up. If you bogged down at 90:1, maybe it's worth looking at tweaks to the server or network. Remember that at this point, Citrix is hoping to make a pile of money from you, so they may be willing to lend engineering help to your cause.
Reading and research is good, but not sufficient. Your environment is like nobody else's except perhaps your closest competitor -- and they're not going to help you out. You need to test YOUR setup and YOUR users.
Forward, retransmit, or republish anything I say here. Just don't misquote me.
Citrix is expensive, and Terminal Services on Win2K and Win2K3 is very, very good.
Do you NEED Citrix? or are there other applications out there that does what Citrix does at a lower cost? Look at Newmoon Systems they have some excellent solutions that are very cost effective and superior to Citrix in environments.
Call you IBM Direct representative. They have some excellent whitepapers regarding server virtulization and Citrix (from a company called Conseco Finance I believe)
Do your research and talk to LOTS of people. Maybe even hire a professional consultant to help you with this project.
My office uses a combiantion of Thin and Fat Clients in carefully monitored and locked down environments to provide applications to 30 medical offices in 18 states. Thin clients don't do everything and Fat Clients are not always better, but Citrix is an expensive licensing beast.
Good Luck. Write a How-To when you are done!
Rule of Life Number 2: Remember, it can all go to hell at any minute. --Jimmy Buffet
A few points from an arms-length admin (We have a 600+ user MF farm, I don't admin it [happy] ): /dev/null.
- Printing can be a nightmare. The individual Win print spoolers hang, eating cpu and stalling print jobs for all users on that server. The solution to print jobs going just anywhere was to create an initial default printer eqivalent to
- Roaming user profiles are insane. Add another server to collect them; it _must_ be up or the entire farm is useless. Need to make a global change? Everyone loses all customizations after all profiles are reset to the default. Try to find what svr a user is on? Enjoy browsing the uselist on each machine.
- MS apps are the worst-behaved. Heavy memory footprint, cpu monopolization, cross-user flakiness & registry weirdness. What fun!
- Vendors tout MF compatibility, it will be your problem to make it work (at all).
- Install mode, execute mode and uninterruptible install reboots.
The MF admin now takes a fanatically strict approach about what makes it on the farm. Any slight weirdness, it don't make it on.
"Remember, any tool can be the right tool." -- Red Green
If you've ever worked in a help-desk or other support environment, you see that training is not the last word in indirect costs for software decisions. The amount of training required to move to Evolution or OpenOffice is really not significantly more than you need to move users to a new version of the equivalent Microsoft app.
It's not a definite win, and certainly something that would be difficult to do for everyone overnight, but you need your IT staff for something after all those thin clients get rolled out, right?
As an extra note, if you can do this with Linux, OpenMosix becomes an option. This way, the users can all use (to them) a single server which you administer, and the load is distributed fairly evenly throughout the cluster. SSI clusters and centralized management can really eliminate a lot of headaches and provide the most performance benefit (since *Mosix automatically tries to get the best performance for a process) when you have 1000 users to support.
--That's the point of being root, you can do anything you want, even if it's stupid.
Is thin client really a cost-saving approach to a large user environment?
In my experience, yes. Look at this:
At a customer site we have the following implementation, fully Linux: every location has one or more application servers where people log in with their thin clients. There is a master application server. At every location on application master is promoted to be a local master. Every night the local masters synchronize with the global master, and later the remaining application servers synchronize with their local masters.
This means that you have to install an application server only on the global master, the next day all other servers have that application as well. Zero point of administration when it comes to that. The configuration files are not synch'ed from the global master, but at every location the application servers synchronize their configuration (/etc directory with a few exceptions) with their local masters, which means for example that configuring a new printer just has to be done on the local master, the next day all servers at that location know about that new printer.
Users' home directories are mounted via NFS from another server.
And now for the Windows part in this picture: we use VMWare and their persistent disk images (I think they're called like that; they throw away all changes and remain constant). On these VMWares we run Windows 2000 servers to which people can connect via rdesktop (RDP). We are now able to administrate just the Windows image of the global master (by temporarily switching off the persistance option), and the next day all other Windows images are the same. That's also pretty resistant to viruses and worms: just reboot in case of infection :-)
This saves money, since the only points to administrate are the global master and sometimes the local masters (site specific configuration stuff like printers).