So...
We were all told that cloud never goes off line, more flexible and less hassle. Now we're being told that cloud is fallible, and you should look to insurance to cover your losses if when cloud goes off-line.
Hmmm.....
Move to the cloud, they said, everything will be better, they said. Security, reliability, scale. We take the work and the worry off your hands. Except nothing is that simple or straight forward – and that includes cloud. When your IT ran the tin and it crashed, they weren’t running your entire business. If a server flamed out …
> it seems you can fool some of the people all of the time
That reminds me of the saying "You can fool some people all of the time, or all people some of the time, but you can't fool all of the people, all of the time", at least how I heard it.
A saying that seems to be more and more true the more times I experience things in life.
I was at an outfitter shop one day, listened in on a conversation about a roof rack.
Salesman demonstrates why a certain (expensive) rack is the best because it's so easy to just open the clips and remove it when not needed. Why drive around with a rack when you don't need to, only a few seconds and it's gone from the roof.
Customer says that sounds very good.
Then salesman points out that a thief could also easily open those clips and make off with it, so for only $40 extra per lock he could buy 4 locks to protect each clip.
At this point I drifted off, no need to spoil lunch.
In my experience managers often push business decisions that benefit their own standing. It has been called empire building when it was restricted to internal staffing and facilities but it cn equally cover management of external contracts and expenditure. It may noit be right for the company (this article is an excellent example).
Business decisions of this importance should carry personal risk to the senior managers who wil benefit from them:
Chairman/CEO to IT Manager and CFO : "You are proposing that we put business critical functions into the Cloud. Are you satisfied that the reliability and cover is there?"
IT/CFO "Oh Yes!"
Chairman/CEO "You realise that if there is a business affecting failure it will be your successors' first task to end these contracts and bring the functions back in-house. The Business financial loss will be placed on your bonuses and pension funds"
IT/CFO "I think we need to check it again. We'll get back to you"
> And when it is the CEO pushing for it, and the IT/CFO team against all things cloudy?
Then the board will rip the CEO a new one. If the board pushes for it (I don't see why they would, it is not within their purview) and it crashes and burns, well it was their money (or the shareholders, which can sue the board if there is a case to answer).
At least that is how it is supposed to work, admittedly it doesn't always.
Chairman/CEO "You realise that if there is a business affecting failure it will be your successors' first task to end these contracts and bring the functions back in-house. The Business financial loss will be placed on your bonuses and pension funds on employee wages and benefits"
IT/CFO "<I think we need to check it again. We'll get back to youSure, NO PROBLEM!"
And if you keep it in house, then who are you gonna fire and fine when it all goes TITSUP? Or are you only advocating punishing managers who want to make changes?
A decision not to change is also a decision. If the IT/CFO is answerable for the result, then that needs to apply whichever way they make the decision. And of course failing to make it at all is always wrong.
Chairman/CEO: "I think we need to check it again. We'll get back to you."
You make a profit in the Insurance industry by selling a "product" that you never deliver.
If you must buy insurance then always read the contract from the back first. The front of the contract always tells to what they cover. The pages at the back of the contract will explain why the items prominently mentioned at the start of the contract are not covered.
But cloud providers are unlikely to take on business impact liability based on the value of a customer’s data, because it’s a difficult thing to measure and could potentially cost them dearly.
So the best thing to do (if you MUST use the cloud) is to HA your business between two cloud providers, and replicate your data.
Your data your responsibility. Simples!
where you agree that you can't sue [insert cloud provider name here] for anything.
Just list these points when presenting the case against it to your PHB
1) They have you and your business by the short and curlies.
2) They'll (if they are still around) increase the cost well over inflation and exchange rates year on year
3) Good luck moving your terrabytes of data to a different supplier
4) You won't have any staff that know your business left to even start to pick up the pieces when it all goes TITSUP as it surely will.
The PHB will blindly ignore all that and after giveing you your P45 they'll engage a few Monkeys from one of the Indian (Cough, cough) Consultancies at 10% of your cost just to show the BOD that they have support in place.
Happy Christmas.
I have to look at cloud provision on behalf of clients. Decoding what service is being offered takes ages. As may be expected the terms are mostly disadvantageous to the clients. The location of the services and support are often obscure, the level of support is always obscure. There's one large provider I wouldn't touch with yours. Organisations are at risk of falling foul of Article 25 of Directive 95/46/EC, simply because cloud providers choose to sling the data around for their convenience. Often the people buying the services don't have a clue what their legal obligations are, or care.
I cite here the immortal statement from Diana Mary "Dido" Harding, Baroness Harding of Winscombe on the Today programme "No one told me that I had to look after this data."
There's a simple way to sum up cloud provision.
...AND the mitigating "technical solutions", as alluded to in the article. Such as the backup the vendor doesn't actually provide or charges several body parts for. Such as inter-site replication (if they offer it).
It all adds up, and frankly for apps that are used by all staff, such as email, particularly if there is a high service expectation, I don't think it makes sense once you get into medium-sized enterprises, unless you literally have no on-prem IT at all. Or if the on-prem IT is insecure and unreliable.
Sure, use the cloud to supplement your backup solution. Use it for apps that are not business-critical or have a small user base. For what remains, BE CAREFUL.
Once upon a time, when working for a large cloud provider who is a shade of 0x0000FF, I noticed something when setting up the systems: the service uptime is dependent on how the user provisions and configures their hosts. You have to configure hosts in separate regions, with the data replicated between them. At least two, preferably three regions. Oh, and you're billed for the data transfered.
So there you have it: your expenses are x3, plus data transfer. Now you're safe from somebody mis-configuring a switch and blacking out a region, and all your money are belong to them.
Is this a fake news site?
The whole premise of this is ludicrous: don't bet your business on the cloud because it might go down. It will. Will it go down more than if you run it yourself? Maybe, maybe not. Probably not. A lot can go wrong that's not the providers problem that orgs consuming cloud service must "own" - even if you are your own onpremise provider. Here's one view of this: . 5 Things that Can Go Wrong with Cloud networking ( https://www.exoprise.com/2015/11/10/5-things-wrong-office-365-networking/ )
Here's one view of this: . 5 Things that Can Go Wrong with
Quite a bit of advertising you're doing for yourself there.
But tell me. Do you know the differences between the meanings of the words "premise" and "premises" ? Because those two words have quite different meanings, and for someone who is a provider of anything relating to communications (network or otherwise), a grasp of basic English is a must-have. If you cannot even choose to use the correct option of "premise" and "premises" then there's a good change that your product is also not the correct choice.
Next time you want to use El Reg to advertise your product, I suggest you do the right thing and take out a paid ad.
If you manage your own servers in-house, a serious failure can screw your entire business for much more than just a day.
The "compensation" clause in the service contract isn't about compensating you, it's about aligning your provider's interests with yours. If my cloud service goes down for as long as 24 consecutive hours, the compensation will wipe out the provider's entire year's profit from my business. Ergo, they are strongly motivated to make sure that doesn't happen.
I work with on prem and cloud based solutions equally and sell and support both. I have do inherent pull to one or the other as their both, in the end, tools to support a business.
But when we look at cloud-based services, the cost is seen as quite cheap. What are you really expecting for that? If you are paying, say, $30 per user, per month, then you are paying ~37 cents for 9 hours (8:30a-5:30p)
If that's the value you put on that then so be it.
It's not a matter of right or wrong, just about expectations. If you payed $37 for your company of 100 people to use something for a day, what compensation can you really expect if that falls through?
I'm not saying that this is acceptable and that cloud providers shouldn't be more accountable but that's the reality.
It's easy to point and laugh at the spectacle that results from outages of major "cloud" services (let's not kid ourselves, it's a lot of fun, too!). However, I wonder how their reliability compares to "in-house" I.T. / phone services.
I escaped the I.T. sector several years ago (which was fortunate for my sanity and the safety of everyone around me). I remember "back in the day" that our in-house I.T. and phone systems experienced complete outages from time to time, at a frequency of about 2 to 4 times per year, on average (per my best estimate).
[For the naysayers: no, I was not responsible for setting up or administering these systems.]
Moving to the cloud so that the NSA and CIA and FBI and anonymously accessed by the mossad means no more IP security.
To all your engineers please make sure you time stamp everything so that when you invent something that later it does not show up in China or Israel or Russia. All 3 countries are the top famous countries to hack and steal info so if you suffered IP theft before then DO NOT rely on the CLOUD at all!