How often do you hear - “Does Cloud really save money?”

[Tefters]

Again it can't just be me sick to the back teeth of being proactive and not being acknowledged for it, nor given any more budget as a result. Those cost savings and improvement should be recognised. In private sector and industry you used to get bonuses for that.

And here is the reason why I am looking at going back to the private sector, I love education and feeling like you making a difference to these kids lives and keeping the school in the modern era while making teacher lives easier however when teachers keep getting pay rises and bonuses for performance year on year and yet the best us IT admins/managers are seeing is a pat on the back what's the point...

I can go and earn more money working as a technical consultant quite easily or even infrastructure engineer with less responsibility.

 

8 years I've been working at my establishment with 0 relevant IT training other than what I learn/study myself in my own time and now mater how much I ask for a training budget its always "email numbers and we'll take look" and then never do.

 

In short - Everything other than teaching is just irrelevant.

 

All off topic - Apologies!

Edited April 5, 2019 by Tefters

[PotNoodleTech]

I have not heard “Does Cloud really save money?” very often.

 

What I have heard often is "Cloud saves you load of money weeeeeeeee"

 

And if you do some real maths, no it doesn't. Cloud only saves you loads of money if a) your systems are all out of date and need replacing, and b) your paying through the nose for enterprise solutions that have no place in a budget constrained school.

[localzuk]

Cloud saves you money just like VDI saves you money. In very specific circumstances, carefully planned out and any variation away from that plan ruins the cost savings.

[DrCheese]

Ah yes, VDI ! I know of one large estate agent that went full VDI, only to reverse it a few years later as the downsides meant they kept having to put in variations that ultimately destroyed the projected cost savings. Costs of putting in ultra fast broadband in some of their more remote offices really bit hard.

[stgoodyeara]

That DfE report is the BIGGEST load of waffly spherical objects I’ve heard this year... Yes! I am comparing that to the BrexIt garbage as well.

 

‘Cloud First’ approach - You mean the new latest buzz term and jumped on the band wagon to look like modern and hip...

 

Allowed practitioners and school leaders to focus on what matters, teaching and learning - REALLY?!?! Since when have SLT cared about IT unless it’s broke or they don’t have all the toys they want.

 

Automating transfer of assessment data between 2 systems - Yeah... because that’s cloud only...

 

Saving of 50% - Compared to your old archaic setup that was costing you a fortune because no-one was proactively looking at new solutions each year to save money... For the last 6 years I’ve saved huge amounts year on year thanks to new tech and reinvested into further new techs to make life easier, more efficient and slim line.

 

Cloud DOES NOT save you money majority of the time when comparing solutions properly. It’s the new lazy ‘outsourcing’ method and I can’t wait to see what people start saying when those costs start ramping up.

 

What about your primary and secondary schools in villages with only ADSL connections? Or secondary’s paying £8/9k on 100/100 lines which are already heavily used for normal use let alone offloading semi-file servers etc...

 

Cloud is being advertised and pushed by everyone like Apple marketing and it’s stupid.

Yes cloud does have its place and in some MAT’s / new schools it might be worth it but to issue statements of ‘Cloud First’ or ‘Cloud Future’ is stupid and laughable.

 

Couldn't agree with these statements more! Im in rural north yorkshire and 200/200 is around 8k per year and we are iPad 1:1 so there is NO WAY id dare adopt anything cloud other than out of hours backups, our lines are heavily subscribed all day long.

 

A big NAS in a different outbuilding is a large initial expense but doesn't have the year on year costs of cloud. Plus if you ever want to restore data from the cloud....your often paying again.

 

Plus in a disaster recovery situation i can restore data faster from my multi NIC'd NAS box than i can download it over an already heavily subscribed data line in the middle of a school day. So to make cloud workable i'd need another £8k data connection which is a annual cost of more than the initial outlay for a decent NAS box.

 

The people that write these articles need to base arguments in the real world, and when they discover they can't maybe shouldn't publish

[Ditto]

I have not heard “Does Cloud really save money?” very often.

 

What I have heard often is "Cloud saves you load of money weeeeeeeee"

 

And if you do some real maths, no it doesn't. Cloud only saves you loads of money if a) your systems are all out of date and need replacing, and b) your paying through the nose for enterprise solutions that have no place in a budget constrained school.

I'd suggest 80% of schools fit criteria a) and b) and I think these forums can provide many examples to illustrate this.

[HPlum78]

Broadband connectivity in any school (or business) is a sunk cost these days you can no longer identify this as something that can be sacrificed just like your water and gas (speed of connectivity is a different matter)

 

This risk that your all talking about of the supplier putting the price up really, MS could put their on prem licensing costs up in fact when they got hit by the EU fine for IE they put the price up to the exact cost of that fine. VMWare they are not known for price hikes are they! So don't tell me that on prem solutions are any more reselient to price increases than the cloud.

 

I am not saying that the cloud is THE defacto solution either its somewhere in between. Stop running on prem hardware and start being transformational in the cloud or on prem.

Edited April 5, 2019 by HPlum78

[Field_IT]

There is a saying that, There is no such thing as Cloud, it is just someone elses computer!

 

Just saying

[CloudGeek]

There is a saying that, There is no such thing as Cloud, it is just someone elses computer!

 

Just saying

 

Yes, someone else’s computer, absolutely.

 

However with well designed true cloud application (not legacy client-server Windows thick client application), all resouurces (CPU, memory, IOPs) are shared amongst all users of the same application. And this is where saving comes in.

 

Otherwise you are right, hosting SIMS7 means you physically move on premise server to a hosted location as “someone else’s server” rather than yours.

 

And yes, with true cloud MIS you will save shed loads of money - up to 50%.

 

If don’t believe it why not ask DfE or Ark Schools to share the details. If they claim this then they must have facts to support it!

Edited April 5, 2019 by CloudGeek

[DrBeaker]

There is a saying that, There is no such thing as Cloud, it is just someone elses computer!

 

Just saying

 

Yep and they can increase charges as much as they want or even shut the service down. To those who think this doesn't happen, see the below articles.

 

https://www.bbc.co.uk/news/technology-47810367

 

https://www.droid-life.com/2019/04/05/artist-hub-for-google-play-music-shuts-down-on-april-30/

[mukz]

And yes, with true cloud MIS you will save shed loads of money - up to 50%.

 

If don’t believe it why not ask DfE or Ark Schools to share the details. If they claim this then they must have facts to support it!

 

I put this down as economies of scale personally, a large MAT will save money by consolidating services.

 

We are moving to the cloud and we have saved some money by ditching some stuff onsite but it only works out cheaper over a period of 10 years because we are not having to renew servers.

[CloudGeek]

I put this down as economies of scale personally, a large MAT will save money by consolidating services.

 

We are moving to the cloud and we have saved some money by ditching some stuff onsite but it only works out cheaper over a period of 10 years because we are not having to renew servers.

 

Economies of scale does come in because Microsoft buy servers for Azure by tons! But this extra. The fundemental saving comes the fact that properly designed cloud application share same instant of code by number of users simultaneously and number of users limited not by memory by the CPU. Because each extra user takes fraction of extra memory comparison a Windows application. In summary resource sharing is where the saving comes from.

 

For sharing, you do not need to be within a MAT. Well designed cloud application will allow unconnected users secure total privacy.

 

I shared this short video before under different thread but here it - Phil Wainewright -True v Fake Cloud - explained in a simple language:

 

I hope it helps.

[daustin]

Cloud does have some benefits over on prem but it really really depends on circumstances, uses and cost. I think we could all possibly find some uses for cloud in a small/big way and many probably have with o365 and GApps but its certainly not a case of every school must go cloud because "woo cloud"

 

CloudGeek I notice that video is a paid promo for Bromcom, just to confirm you dont have any connection to Bromcom? be it paid staffer or even advocate for them?

I just get very suspicious when I see promo video's appear for an MIS provider.

 

Some of the video makes sense but being cloud based doesnt mean if it isnt a web page its not real cloud, cloud is much much more than that.

[Steve21]

CloudGeek I notice that video is a paid promo for Bromcom, just to confirm you dont have any connection to Bromcom? be it paid staffer or even advocate for them?

I just get very suspicious when I see promo video's appear for an MIS provider.

 

Even if the answer is said to be no, basically seems so yes.

 

Every single one of these recent MIS cloud threads and posts is the same thing, a bunch of drivel and twisted stories linked to one so called advertiser to get people to change to their product. Look at the post history, 49 posts and 49 are slagging off Capita and suggesting moving to "insert brand here", doesn't that say it all?

 

Really getting sick and tired of these threads that bring no information or backup data that's being shouted about, and yet as soon as it's questioned turns into again the same company being advertised for. Just like this so called "MIS alliance" that keeps popping back up again and again: http://www.edugeek.net/forums/mis-systems/204420-foi-request-re-upgrading-capita-sims7-cloud-based-version.html or a dozen other examples that magically even the official PR team disappeared on when replying... Says a lot really

 

I have nothing against cloud services, but again it's just the fact that if there's no set evidence or data to backup these claims it really shouldn't be shouted about...

 

Steve

Edited April 6, 2019 by Steve21

[mavhc]

You've paid for 24/7 fiber line that you're not using at night, can you leverage that cost to save money?

 

1. Do you have >1 site? Yes. Then you can do offsite backups to another site. No: probably look into cloud backup.

 

2. What is your backup for?

2.1 Site destroyed, need complete backup, and a new site.

2.1.1 Can you run your backup from the cloud, start a VM etc.

2.1.2 How long will it take/cost to restore.

2.2 Need a copy of an old file, how fast is that to achieve?

2.3 Hardware failed locally

2.3.1 Software RAID, multiple failover servers etc. Whether that's worth it depends on your size. Small school with 1 server won't have budget for 2. But large school with 10 will have budget for 11.

2.4 Malware encrypted stuff

2.4.1 How do you ensure the encrypted version isn't backed up? Or the old version is still around.

 

3. Cost of backup hardware/software. Apart from the drives this should really be almost free. It's almost too easy to snapshot a vm and copy it. Buy a £500 24 drive case and computer, fill it with 8TB nas drives, install zfs, backup to it.

 

4. Moving applications to a cloud. Either you a) start paying for a subscription to someone's SaaS, or you b) create a container you can move to any cloud system you wish, if one changes prices you just move to another, or back onsite, very flexible. Save money by scaling it up and down as required, turn it off at night etc

 

Problem with PaaS is lock in, price increases.

[djrscally]

Backing up your on prem data to the cloud is an interesting one and its not as straightforward as a direct comparison. You have to have a good reason for considering architecting a solution like that and in the most part it's not cost effective the ingress and egress costs alone make it too expensive for most.

Backblaze is $0.005 per GB per month; it's not that expensive to backup data to the cloud.

 

And yes, with true cloud MIS you will save shed loads of money - up to 50%.

 

This just is not necessarily true. The running costs of 500 cloud instances of an MIS may well be lower than the running costs of 500 local installations; that doesn't mean that a *customer* is going to see those cost savings. It depends entirely on provider pricing and there are cloud MIS's that easily match SIMS on pricing - with the hosting part of the bill being way more than it would cost for a server spread over 10 years or so. There are also significantly cheaper ones it's true, but there are likewise cheaper locally installed alternatives.

 

My view in general on cloud hosting is that it is of absolutely invaluable benefit to particular functions, but has become such a profitable buzzword for salespeople that completely inappropriate things are now shoehorned into a cloud environment. In the particular case of a school MIS it's swings and roundabouts. If you lack staff with particular skills then they're a safer choice. If you have those staff then I think they're a marginally poorer choice. I have seen no evidence that there is a cost benefit or detriment either way.

 

Maybe we need to run a Maloney-style FOI to every school to find out what MIS they use and how much they pay for it.

[Ditto]

Maybe we need to run a Maloney-style FOI to every school to find out what MIS they use and how much they pay for it.

 

Perhaps we don't need it Maloney style! It would be a really useful addition to this analysis. I'd suggest a cost per pupil would be informative.

[Ditto]

You've paid for 24/7 fiber line that you're not using at night, can you leverage that cost to save money?

1. Do you have >1 site? Yes. Then you can do offsite backups to another site. No: probably look into cloud backup.

2. What is your backup for?

2.1 Site destroyed, need complete backup, and a new site.

2.1.1 Can you run your backup from the cloud, start a VM etc.

2.1.2 How long will it take/cost to restore.

2.2 Need a copy of an old file, how fast is that to achieve?

2.3 Hardware failed locally

2.3.1 Software RAID, multiple failover servers etc. Whether that's worth it depends on your size. Small school with 1 server won't have budget for 2. But large school with 10 will have budget for 11.

2.4 Malware encrypted stuff

2.4.1 How do you ensure the encrypted version isn't backed up? Or the old version is still around.

3. Cost of backup hardware/software. Apart from the drives this should really be almost free. It's almost too easy to snapshot a vm and copy it. Buy a £500 24 drive case and computer, fill it with 8TB nas drives, install zfs, backup to it.

4. Moving applications to a cloud. Either you a) start paying for a subscription to someone's SaaS, or you b) create a container you can move to any cloud system you wish, if one changes prices you just move to another, or back onsite, very flexible. Save money by scaling it up and down as required, turn it off at night etc

Problem with PaaS is lock in, price increases.

 

This is a knowingly slightly blinkered perspective, but in the case of the GSuite and Chromebook environment I worked in...

1. Do you have >1 site? Yes, about 4 sites, but we relied on Google's backup and security.

2. What is your backup for? Accidental deletion. The files stick around quite a while even when deleted from 'Trash', but I guess we could archive elsewhere.

2.1 Site destroyed, need complete backup, and a new site. Provided we could get a internet connection, we were nearly 100% functional. I guess we'd need to source new CBs, but a PC would also allow most functions to work.

2.1.1 Can you run your backup from the cloud, start a VM etc. Yes, and we didn't have VMs

2.1.2 How long will it take/cost to restore. In terms of time, it would come down to having replacement classrooms with internet. Costs - hard to know but would expect fully covered by insurance.

2.2 Need a copy of an old file, how fast is that to achieve? Most Google documents have a running file history - a matter of minutes.

2.3 Hardware failed locally. We'd have spare CBs. The rest would be most AP's and core network - the same whether cloud or internally managed.

2.3.1 Software RAID, multiple failover servers etc. Whether that's worth it depends on your size. Small school with 1 server won't have budget for 2. But large school with 10 will have budget for 11. Not an issue.

2.4 Malware encrypted stuff. Maybe because of ChromeOS, we never hit that issue, but for us not a cloud/non-cloud issue.

2.4.1 How do you ensure the encrypted version isn't backed up? Or the old version is still around. Not an issue.

3. Cost of backup hardware/software. Apart from the drives this should really be almost free. It's almost too easy to snapshot a vm and copy it. Buy a £500 24 drive case and computer, fill it with 8TB nas drives, install zfs, backup to it. Free - there's a cost per device for device management, but assuming a 4 year life of device, let's say £5 per annum per CB.

4. Moving applications to a cloud. Either you a) start paying for a subscription to someone's SaaS, or you b) create a container you can move to any cloud system you wish, if one changes prices you just move to another, or back onsite, very flexible. Save money by scaling it up and down as required, turn it off at night etc. We paid for some Chrome apps/extensions, but found prices competitive on the whole.

 

And perhaps this is the tough part. We started from scratch with Cloud based solutions. Whilst some traditional practices crept in, we didn't have to manage a migration away and I can see that's a challenge. I'm not daft enough to think Google and a good internet connection is the utopian solution for all, but I do feel there is real potential, but you have to go in open-minded, with a positive attitude and do the groundwork.

[HPlum78]

@djrscally I had discounted backblaze as they don't have an operational EU/ UK (you should really consider what you are storing outside of the EU) data centre so are at the moment discounted for being a target to backup to, also they have not firmed up what the pricing will be as they have openly stated its going to more expensive. Edited April 6, 2019 by HPlum78

[Arthur]

also they have not firmed up what the pricing will be as they have openly stated its going to more expensive.

Didn't they already increase prices last month?

 

 

https://www.backblaze.com/blog/backblaze-computer-backup-pricing-change

[HPlum78]

Sorry more expensive than the US data centre prices.

[djrscally]

@djrscally I had discounted backblaze as they don't have an operational EU/ UK (you should really consider what you are storing outside of the EU) data centre so are at the moment discounted for being a target to backup to, also they have not firmed up what the pricing will be as they have openly stated its going to more expensive.

well ok that's fair enough if that's a concern, but AWS and Google both do have EU options for only marginally more money, like $0.007 instead of $0.005 I think.

 

Didn't they already increase prices last month?

 

 

https://www.backblaze.com/blog/backblaze-computer-backup-pricing-change

 

 

These are for their specifically backup thingies. They include, I believe, a local client to do the scheduling, compression, encryption and transmission to their servers. I have never really understood why people might buy those; you can just back up and encrypt locally with one of the several available open source tools and transmit the data to their B2 cloud storage, which is dirt cheap.

[mavhc]

$0.005 is $60 or £40/year for 1TB, which is great if you have 1TB to backup.

 

But HDDs are £25/TB and last 5 years, plus £5 in power, so £6/TB/year.

 

Cloud scales linearly, but DIY is much better.

 

 

£100 2 disk nas for up to 8TB.

 

£500 24 disk case for 160TBs (2 12 disk RAIDZ2), £100 for motherboard, £100 cpu and ram, 3x £50 HBA, £200/8TB HDD

 

Assuming you have a 2nd site to store the data.

[djrscally]

$0.005 is $60 or £40/year for 1TB, which is great if you have 1TB to backup.

 

But HDDs are £25/TB and last 5 years, plus £5 in power, so £6/TB/year.

 

Cloud scales linearly, but DIY is much better.

 

 

£100 2 disk nas for up to 8TB.

 

£500 24 disk case for 160TBs (2 12 disk RAIDZ2), £100 for motherboard, £100 cpu and ram, 3x £50 HBA, £200/8TB HDD

 

Assuming you have a 2nd site to store the data.

 

Yeah if you have another site I'd go for that option too.

[mavhc]

Can always make a pact with the next school over. ZFS 0.8 adds built in encryption