Microsoft to move out of the education market?

[localzuk]

As a % of your schools turnover what do you think the price "hike" works out at?

 

For us it works out around 0.017%

 

I'm not going to lose any sleep over it....

 

You must be a well funded school then.

[kennysarmy]

You must be a well funded school then.

 

Quite the contrary - we're one of the schools that don't come off very well regards the funding anomalies around the various counties. But we are very well run financially.

 

Anyway, if I round our turnover up to ~ £7,000,000 and take the price hike of 19% on our last years annual MS licence costs at around £1400. Then £1,400/£7,000,000 * 100 = 0.02%

 

Whilst I appreciate the budgets are tight in education I don't see the price hike being something that should be forcing Network Managers hands!

[gshaw]

As a % of your schools turnover what do you think the price "hike" works out at?

 

For us it works out around 0.017%

 

I'm not going to lose any sleep over it....

 

At our FTE count the final 20% hike works out to be around £4k, far from insignificant in the current climate.

 

I'm not convinced that'll be the end of it either as MS will have to force people to move certain parts of on-prem to Azure as there's very little reason to do it otherwise.

 

Office 365 email, Skype etc are great in the cloud and makes a lot of sense (huge storage, easy management etc) and many have embraced this offering.

 

Existing legacy apps, domain controllers, print servers etc. moving to a subscription charge in Azure, not really compelling.

[localzuk]

Ah, I see where your difference is. We're a middle school. so get less funding per pupil overall...

[CyberNerd]

Existing legacy apps, domain controllers, print servers etc. moving to a subscription charge in Azure, not really compelling.

 

This won't be the case in a few yrs.

[Ephelyon]

I don't know if that's true - I'm not sure when the business logic would ever apply that it makes sense to transmit a document from a workstation to a print server over the Internet for processing, then transmit it back down the line again to a printer on the local network. It's a clear example of a workload that should definitely remain on-site where you have the choice, especially as the connection bandwidth you do have will be under more strain from all your other cloud workloads by that point than it is today.

[localzuk]

I don't know if that's true - I'm not sure when the business logic would ever apply that it makes sense to transmit a document from a workstation to a print server over the Internet for processing, then transmit it back down the line again to a printer on the local network. It's a clear example of a workload that should definitely remain on-site where you have the choice, especially as the connection bandwidth you do have will be under more strain from all your other cloud workloads by that point than it is today.

 

Ah, but that's the cloudy thinking - you don't have a print server in the cloudy world. You use cloud enabled printers instead. It goes with the idea that every person will have a device too, so won't need to print, so there'd be no need for limiting the number of prints that can be completed.

 

Its all part of the "paperless office" idea. Which is why this office has 11 filing cabinets in it.

[CyberNerd]

I don't know if that's true - I'm not sure when the business logic would ever apply that it makes sense to transmit a document from a workstation to a print server over the Internet for processing, then transmit it back down the line again to a printer on the local network. It's a clear example of a workload that should definitely remain on-site where you have the choice, especially as the connection bandwidth you do have will be under more strain from all your other cloud workloads by that point than it is today.

 

You are not transmitting from a workstation to the internet - the work is already on the internet. I agree that it is downloaded if it is printed, but it's usually in a compressed format. All of our printers work on Google cloud print and it's barely affected bandwidth I don't think it would make the slightest difference if the servers were hosted on site or in GCE. Here bandwidth is more affected by windows update and youtube - printing barely registers.

[localzuk]

You are not transmitting from a workstation to the internet - the work is already on the internet. I agree that it is downloaded if it is printed, but it's usually in a compressed format. All of our printers work on Google cloud print and it's barely affected bandwidth I don't think it would make the slightest difference if the servers were hosted on site or in GCE. Here bandwidth is more affected by windows update and youtube - printing barely registers.

 

You've not seen some of the things we print then...

 

Printed a 700Mb booklet yesterday.

[CyberNerd]

You've not seen some of the things we print then...

 

Printed a 700Mb booklet yesterday.

 

what was it made in ?

[localzuk]

what was it made in ?

 

Indesign.

[CyberNerd]

Indesign.

 

we used to see this with publisher in the olden days.

[localzuk]

we used to see this with publisher in the olden days.

 

Complex DTP projects with high quality imagery take up lots of space. Its part and parcel of the process. Especially if you then want to get that project printed professionally.

[CyberNerd]

Complex DTP projects with high quality imagery take up lots of space. Its part and parcel of the process. Especially if you then want to get that project printed professionally.

 

sure but this shouldn't be part and parcel of student work. If it's going to a professional printer, share the link to them and use their bandwidth!

 

Anyhow - it seems it's our sixth form windows BYOD that use our bandwidth:

 

Top domains (24 hours)

tlu.dl.delivery.mp.microsoft.com 14.1 (GBytes)

au.v4.download.windowsupdate.com 6.3 (GBytes)

Google 3.5 (GBytes)

s3.amazonaws.com 2 (GBytes)

iosapps.itunes.apple.com 1.5 (GBytes)

umgstore.edgesuite.net 1.5 (GBytes)

as-hds-uk-live.edgesuite.net 1.5 (GBytes)

aupl.v4.b1.download.windowsupdate.com 1.4 (GBytes)

hls-aka.vevo.com 1 (GBytes)

upload.wikimedia.org 1 (GBytes)

 

kind of ironic since we use GAFE exclusively.

[localzuk]

sure but this shouldn't be part and parcel of student work. If it's going to a professional printer, share the link to them and use their bandwidth!

 

Not really how design work is done... Lots of mock ups, lots of prints until you're happy with it. Does apply to student work too, as well as staff work.

 

Anyhow - it seems it's our sixth form windows BYOD that use our bandwidth:

 

Top domains (24 hours)

tlu.dl.delivery.mp.microsoft.com 14.1 (GBytes)

au.v4.download.windowsupdate.com 6.3 (GBytes)

Google 3.5 (GBytes)

s3.amazonaws.com 2 (GBytes)

iosapps.itunes.apple.com 1.5 (GBytes)

umgstore.edgesuite.net 1.5 (GBytes)

as-hds-uk-live.edgesuite.net 1.5 (GBytes)

aupl.v4.b1.download.windowsupdate.com 1.4 (GBytes)

hls-aka.vevo.com 1 (GBytes)

upload.wikimedia.org 1 (GBytes)

 

kind of ironic since we use GAFE exclusively.

 

How many devices is that? Our general internet usage uses more data than that at the moment!

[CyberNerd]

How many devices is that? Our general internet usage uses more data than that at the moment!

 

rough breakdown is:

1200 Chromebooks

Probably 350 BYOD (OSX/Chromebook/Windows Mix)

700 phones (guess)

200 Thin clients

80 Macs

270 Windows clients (WSUS!)

 

roughly 2000 devices

[gshaw]

This won't be the case in a few yrs.

 

This is the interesting part, will it be the case because

 

a) the technology moves on and becomes compelling in the same way Office 365 has taken over from on-prem Exchange

b) the supplier makes sneaky licensing changes to make existing infrastructure unviable

 

If it's the second option then Google will be celebrating. There's a big difference between moving to the cloud for services because they're better than what we're doing now and moving because licensing says so.

 

Already been discussing this to some extent with Jisc regarding cloud-hosted MIS, Finance systems etc. Right now there's a lack of products that you'd call "cloud first" that are built to be scalable SaaS systems including proper APIs for data exchange \ integration. Slapping a VM of a traditional on-prem product in Azure \ Amazon and calling it "cloud" is exactly what doesn't benefit anyone except those providing the hosting.

 

In the meantime we have a few more years before this kicks in so be interesting to see what technology brings us by then.

Edited April 28, 2016 by gshaw

[Ephelyon]

Also, even if the document is stored online, owing to the way the printing subsystem works, the data is transmitted by the workstation to the print server (which is necessary for many reasons, one of which being authentication), so that would involve the data coming down the line to the workstation and then back out again to the cloud-hosted print server VM. It won't be the case that the workstation simply instructs the server hosting the document on e.g. OneDrive to access an Azure VM in another area of that cloud directly for the data transfer. It's the same reason why, with file shares, it's best to copy from the machine you're logged on to from a remote one - if you're logged on to your workstation and you copy from one remote share to another, the traffic will come via you and it will take longer.

 

Another example of workload I wouldn't want in the cloud for the same reason is Impero (or similar). When you monitor screens (which of course you can do in bulk), the data goes from the workstation to the server and then back to you (they have confirmed this). Do I really want the bandwidth usage of multiple internal workstations' screen monitoring to go from them, to the cloud and then back to my workstation, as it would be if the Impero Server VM were cloud-hosted? Or would I prefer all of that to stay on the same internal network with far less latency - key word there - so things aren't chugging around through Internet pipelines that don't need to be?

 

I think you're right in that it will be possible to run just about anything from the cloud in a few years' time, but I don't think that means it will ever be the best choice for every conceivable workload.

[gshaw]

I think you're right in that it will be possible to run just about anything from the cloud in a few years' time, but I don't think that means it will ever be the best choice for every conceivable workload.

And that's the key point, same as a few years back when BSF was suggesting to run "everything" via thin clients. The sensible choice is a mix of technologies selected to best fit the use case.

[Ephelyon]

There's a big difference between moving to the cloud for services because they're better than what we're doing now and moving because licensing says so.

 

^ This. I think that whoever we end up sending off to see M$ in May, and whatever else is discussed, we should agree to put forward a collective position along precisely these lines, and make a fairly big thing of it.

 

Who would agree?

[CyberNerd]

 

a) the technology moves on and becomes compelling in the same way Office 365 has taken over from on-prem Exchange

b) the supplier makes sneaky licensing changes to make existing infrastructure unviable

.

 

I think both a) because it's already possible license free with Linux, people just don't have the incentive to learn howto.

b) because the costs of cloud computing are in free fall. Look at the Google Compute Engine costs compared to a few years ago - but people are only now talking about (Azure?) because MS are trying so hard to copy Google.

 

Also, even if the document is stored online, owing to the way the printing subsystem works, the data is transmitted by the workstation to the print server (which is necessary for many reasons, one of which being authentication), so that would involve the data coming down the line to the workstation and then back out again to the cloud-hosted print server VM. It won't be the case that the workstation simply instructs the server hosting the document on e.g. OneDrive to access an Azure VM in another area of that cloud directly for the data transfer. It's the same reason why, with file shares, it's best to copy from the machine you're logged on to from a remote one - if you're logged on to your workstation and you copy from one remote share to another, the traffic will come via you and it will take longer.

 

Another example of workload I wouldn't want in the cloud for the same reason is Impero (or similar). When you monitor screens (which of course you can do in bulk), the data goes from the workstation to the server and then back to you (they have confirmed this). Do I really want the bandwidth usage of multiple internal workstations' screen monitoring to go from them, to the cloud and then back to my workstation, as it would be if the Impero Server VM were cloud-hosted? Or would I prefer all of that to stay on the same internal network with far less latency - key word there - so things aren't chugging around through Internet pipelines that don't need to be?

 

I think you're right in that it will be possible to run just about anything from the cloud in a few years' time, but I don't think that means it will ever be the best choice for every conceivable workload.

 

We *are* running print services in the cloud and as I've been pointing out it isn't nearly as heavy on bandwidth as people believe.

WRT impero: Things like screenshot and streaming desktop data is absolutely minimal - I was running Citrix full desktops on dialup connections on early pentium 15yrs ago

 

The thing your really don't want in the cloud is a caching proxy server! So yes I agree, but not for the points that you made.

[nathan]

Also, even if the document is stored online, owing to the way the printing subsystem works, the data is transmitted by the workstation to the print server (which is necessary for many reasons, one of which being authentication), so that would involve the data coming down the line to the workstation and then back out again to the cloud-hosted print server VM. It won't be the case that the workstation simply instructs the server hosting the document on e.g. OneDrive to access an Azure VM in another area of that cloud directly for the data transfer. It's the same reason why, with file shares, it's best to copy from the machine you're logged on to from a remote one - if you're logged on to your workstation and you copy from one remote share to another, the traffic will come via you and it will take longer.

 

Another example of workload I wouldn't want in the cloud for the same reason is Impero (or similar). When you monitor screens (which of course you can do in bulk), the data goes from the workstation to the server and then back to you (they have confirmed this). Do I really want the bandwidth usage of multiple internal workstations' screen monitoring to go from them, to the cloud and then back to my workstation, as it would be if the Impero Server VM were cloud-hosted? Or would I prefer all of that to stay on the same internal network with far less latency - key word there - so things aren't chugging around through Internet pipelines that don't need to be?

 

I think you're right in that it will be possible to run just about anything from the cloud in a few years' time, but I don't think that means it will ever be the best choice for every conceivable workload.

 

It certainly will be interesting in the future, the cost of broadband is coming down now making a 100mb leased line accessible to primary schools. Will it be a case of leased line costs reducing allowing faster connections and technology improving that use less bandwidth? I understand that may not all happen in the next couple of years, but it is starting to happen.

[drewp]

^ This. I think that whoever we end up sending off to see M$ in May, and whatever else is discussed, we should agree to put forward a collective position along precisely these lines, and make a fairly big thing of it.

 

Who would agree?

 

^ I agree!

 

It will always be a case of what suits the needs of your network/school at the end of the day!

 

The Microsoft Classroom did look interesting, perhaps this could take some of the exsiting content on file shares away from on-premises?

[sonofsanta]

The cost of 100mb leased line has come down... but so has the cost of 10GbE. My LAN will still be a hundred times faster than my internet connection. There's simply no point and no efficiency in shoving most of this stuff out to the cloud.

[gshaw]

I think both a) because it's already possible license free with Linux, people just don't have the incentive to learn howto.

b) because the costs of cloud computing are in free fall. Look at the Google Compute Engine costs compared to a few years ago - but people are only now talking about (Azure?) because MS are trying so hard to copy Google.

 

Both Google and Microsoft copied Amazon if anything. General consensus as I've always read it re: cloud providers for IaaS \ PaaS is Amazon top, MS reasonably close 2nd and all others (Google included) further behind.

 

Don't lump everyone in the "doesn't want to learn Linux" box - we're quite happy with the stuff we run on CentOS but again some products are better suited for that work than others. Also third party solutions and support will affect the choice of what underlying OS is used.