disaggregate means 'to fall apart" doesn't it?
The opportunity for suppliers to join the government's £60m G-Cloud framework is drawing to a close, according to Mark O'Neill, proposition director for innovation and delivery at the Government Digital Service (GDS). "We plan to launch the first tranche of the G-Cloud catalogue in March," O'Neill told the Cloud Expo event in …
disaggregate means 'to fall apart" doesn't it?
And that without cloud-y anything. Amazing, isn't it?
So, what he's really saying is that gov't IT is a jumble of oversold proprietary systems that don't work very well but nonetheless cannot easily be pried apart. And the magic bullet to solve it is to "go cloud".
I can see seamless expansion of capacity while decomissioning the old systems as a valid point, with as only drawbacks paying for the privilege, more depending on networking and thus also raising bandwidth bills, and, oh, making over-expanding and becoming even more reliant on the new jumble with less leeway to get out, even easier. Other than that it's a fine plan. Clearly the bloke subscribes to the "three letters" style of management. As in, the next round of upgrades will only be somewhere after the next bloke steps in and using a new and shiny different buzzword. As a management vision it seems rather limited. Carry on government.
If the government can stop running hundreds of Exchange servers and instead centralise those into the "cloud", then that's great news for us tax payers. There's a nice bit of cost saving which oughtn't be too much of a strain for the departments involved.
What I wonder about really is the multi-million pound projects to put government workers paylips on the intranet, or do all your employee HR functions online, or *shudder* put all of the public's medial records on to some electronically accessible solution. These are all 'custom' applications, which have thousands of hours of consultancy and development behind them. This is also where I suspect the lions share of wasted expense comes from. I can't see this G-Cloud really helping there.
All said and done though, getting anything centralised and cut the duplication of effort has got to be a good thing.
Although I agree in principle (about not running hundreds of exchange servers), centralising them into clouds does not in and of iteself make savings at any point.
First, you are increasing WAN networking load and potentially increasing bandwidth to another party. WAN costs are easilly the most expensive commodity I buy (off framework), and bandwidth is always at a premium. I have three "cloud" options available on current Buying Solutions frameworks, and even with staff and license reductions, none would "save" enough to pay for the bandwidth hike alone, let alone the change costs. Don't think the Audit and Risk groups would approve the move to a system with so many single points of failure for ALL systems either!
Second, we're all at the virtualisation and storage centralisation game now, and x86 hardware is phenominally cheap for the loads we can run. Our DMZ servers cost 1/10th (invoice price, not depreciated) than the 6 year old kit they replaced last year, and are way more powerful and vastly more power efficient. If it's "outsourced", however you figure that, you ARE paying profit margins... the providers are in this for their shareholders, not the public!
But I do agree about the ill-though out projects! THAT is where the real waste lives, breathes and lines corporate pockets.
it then makes for a less appealing single point of utter failure or such a massive juicy target for any passing Orbital Ion Cannons being operated by any number of hostile nations or megla-manical media control groups... looking to make a point.
...responsible for the billions wasted - is that what he said?
I cannot imagine how much his desktop did cost if he's managing to save 80% - obvious where the waste was centred.
"I've not printed anything for 6 months" <--- EITHER He does not work in real world, OR he is dissembling, as I am sure he's used something printed in the last 6 months.
Never buy database software again? Really? I guarantee that Gov will continue to buy database software indirectly in the form of custom systems until AI takes off. I know I will - Disengenuous at best.
It does sound like services that were his responsibility were deeply inefficient and real savings were required... but not necessarily these services - case not proven.
Which brings me to the reporter - SLOPPY WORK: sod "80% cheaper" - means NOTHING (except he must have had really inefficient systems/staff). I want to know what they were / are now costing, not savings percentages that can be fudged for headlines.
Sadly AC is required - I have responsibility for Government ICT budget!
From what i remember of the last govt., somewhere in Mombassa or Mumbai i'd hazzard a guess
Just what about 100 000 UK Gov staff to go.
Now how about *shutting* some of the 100-200 odd data centres HMG has scattered about?
This is nothing more than a power grab. Corporation, and now governement wants to control all your data and they are making it very easy.
Time for action folks.
systemd-free Devuan Linux hits version 1.0.0
Biting the hand that feeds IT © 1998–2017