Can we really use the cloud?

March 31, 2013

One of the things I think about every so often is whether we can make much use of cloud computing (by which I mean external cloud providers, not trying to set up some sort of large-scale university cloud or virtualization environment). Although I may be suffering from a lack of imagination, so far I keep coming up mostly empty.

The problem is that most of our services are internally facing; they're things we provide to our local users. I see the cloud as great for externally facing services (things like websites that the outside world uses) and things that sit behind them to support them, but this is mostly not what we have. Shipping local file storage and so on off to the cloud seems a little counter-productive and also bandwidth (and latency) limited. Something like IMAP email could in theory be hosted in the cloud instead of locally but currently it's entangled with our local file storage; detaching it would have various awkward consequences that would be annoying to the users.

(One argument is that people are already implicitly moving to the cloud as it is, by shifting email to GMail and storing stuff in Dropbox and so on. But even if they are, I don't think we can displace high quality services like GMail and Dropbox with our own efforts.)

One issue for remoting some services off into the cloud is how we deal with interactions with local fileservers and login services. Local fileservers are important for local heavy-duty computing, so how do we bridge them and either remote servers in the cloud using the local fileservice or even remote fileservice (for example if we pushed our web server into the cloud)? I suppose we could nail up IPSec tunnels and the like, but that feels both fragile and worrisome from a security perspective (assuming that NFS even works very well over WAN latencies).

(I'm assuming that issues of cost, network reliability, and any legal problems can all be magically dealt with. This is not necessarily realistic (and some of them definitely are concerns), but the general enthusiasm for clouds basically handwaves them all anyways.)

Comments on this page:

From at 2013-03-31 06:29:28:

I've seen a couple of tertiary institutions that have decided that since students are bringing their own devices, and the need for infrastructure like Computer Labs is lessened then they should try and provide their services to the Internet. The basic mantra being that there is "Internet" on campus and any service is available via the Internet (whither it's on campus or not). A student should be able to work at home, or on campus. Campus might have better infrastructure (eg faster Internet).

Computer Labs, of course, are never going to go away (since some people might not have a computer, and some licensing dictates that you can't just install it on any student's computer that asks), but the general principle that everything should be available via the Internet. They just become machines on the Internet (although probably behind a firewall like you would do for a machine at home). There also may be compute clusters, but they're available from anywhere, and might be replaced with services like ECĀ². Obviously this makes collaboration with other institutions easier too since you can just work with them.

Filesystems like NFS need to be accessible by any student, anywhere on the Internet. Either by Kerberos encrypted NFSv4, something DropBox like, or a system like WebDAV over HTTPS or whatever.

This suggests many things are easier done by throwing it into the cloud (eg Gmail), and some concepts don't really translate that well (eg NFS to lab machines).

Anyway, I thought it was an intriguing concept and perhaps not for everyone, but it does seem to be the other end of the spectrum to what you describe.

--- PerryLorier.

From at 2013-03-31 08:13:29:

According to NIST's definition of cloud computing, one of most important aspects of cloud is apparent infinity of resource you're buying. I think this is the key to finding usage of cloud services. If you expect some service to grow rapidly, you may want to host it on some cloud platform, which would ease expanding resources consumed by the service. Otherwise, I don't see a scenario where outsourcing would help.

-- dozzie

From at 2013-04-02 01:15:55:

I've had similar thoughts about the cloud in my professional work. As an organization we're not ready to move any of our .com sites into cloud infrastructure given how closely tied things are to our datacenters and ISPs. The motivation just isn't there yet.

However I was giving a talk on AWS recently at work showcasing one place we've made cloud computing useful. We have several remote offices which would otherwise be small piles of just 2-3 servers where there are no IT hands present. In this case, we use private clouds from providers like AWS and IPsec tunnels to transparently connect small cloud setups to the local office to provide internal-only services. Why'd we do it? For the infrastructure we host in the cloud, we have 100% remote control of everything from power to machine failures (except if the cloud dies full-stop of course), and I never need to get on a plane to replace a faulty piece of hardware in a faraway country. Except if my (hopefully ultra-reliable) router fails.

Summary: we needed a motivation to use cloud computing, stronger than "it's there." We found a motivating feature - less risk of epic failure requiring expensive travel, and we now have that with our cloud installations.

 --- Patrick van Staveren
Written on 31 March 2013.
« One irritation in xargs's interface
Why we'll continue to have local compute servers in the future »

Page tools: View Source, View Normal, Add Comment.
Login: Password:
Atom Syndication: Recent Comments.

Last modified: Sun Mar 31 00:59:41 2013
This dinky wiki is brought to you by the Insane Hackers Guild, Python sub-branch.