There's only so much you should spend on commodity cloud services before you consider other options like getting a discount, moving back to your own servers, or to a private managed cloud, according to cloud providers and customers. If you're spending more than $10,000 (£6,500) a month or so, then it's probably worth considering …
$10k is not a bad number
If only some of the places I knew had as much sense as these folks.
One company I was at for a while was spending upwards of $400,000/month (at the peak - then came back down to mid-upper $200k/mo).
You see stuff like that and you think "Face wall"
I offered to get them out and save millions, but the board didn't want to. (everyone else did including execs).
I suppose needless to say that particular company is pretty much dead now having axed ~85% of their staff late last year.
I know of a few other places in similar boats - spending six figures a month on cloud while simultaneously not wanting to get out - for stupid reasons too. They know the ROI but they don't care. They believe in the cloud. They bought it hook line and sinker. They experience the outages, the bad performance, the poor support, you name it. After all that they still don't care.
I'm hoping the crappy economy eventually hangs those people out to dry...but am not holding my breath.
Fortunately I haven't had to deal with such stuff in a long time now, stress levels very low now I don't have to worry about amazon crap failing left and right.
Re: $10k is not a bad number
These days, you see stuff like that and you should think: Fraud or Looting!
People on boards are not stupid, they care about what is important to them - Now, if it is the golden parachute, the stock options they have at the service provider, the CDS's they bought on the company's debt or maybe the next, cushier, job at the competitor - who will buy out the present business - That is to be discovered.
Re: $10k is not a bad number
"They experience the outages, the bad performance, the poor support, you name it."
They are using the wrong cloud. I haven't had any of those experiences with Rackspace or Office365.
Re: $10k is not a bad number
There are plenty of places that will give you a 'private cloud' for a lot less than AWS on demand if you sign a contract. http://mediatemple.net/helix/ is a great example.
Heck, even Fujitsu will give you a Windows cloud for better rates than AWS on demand.
"If you're spending $10,000 on public cloud ...there is a cost associated with that"
Wise words there Mark, wise words. Would it be about $10,000 by any chance.
A Stupendously Simple KISS which is Highly Complex and Unerringly True
"Cloud doesn't have to be virtualized," Bailey Caldwell, Rightscale's veep of business development, says.
Cloud Services is always Virtualized …. for Instant Delivery of Presents from Anywhere to Everywhere and Somewheres Nowhere Near Here but All over There. Hmmmm? Heavenly Cloud Services with Definite Vision on AIMissions, ….. Per Ardua ad AstraMetaPhysical Being.*
Almighty Bling to Server Live in LOVEnvironments, which as you may be imagining, has many levels of code and assets capture for mutually advantageous change flows.
The Simple Premise is ….. Words Create and Control Worlds, with IT and Media Mogul Bods Responsible for Presentation of Enhanced Future Realities? Or is that accountability shifted to Virtual Sector Leaderships and is so simply, easily resolved with the daily share of virtual fare for what is to be.
But, no matter, ….. I now go to have a ponder and wonder at future responsibility and present accountability to what and/or whom?
* Crikey, a See with Vatican Hot Lines. :-) Now methinks the pictures there be all bawdy and heavenly to paint and repaint well with satisfaction.
This is a maddening problem that more of us are going to face. Rented cloud infrastructure can get costly fast. Colocating equipment calls for a big commitment, significant more inhouse expertise and unless you have an elaborate stratgegy with fail over spanning data centers, risk of significant downtime if things go south. I like the notion of hosting at least a portion of net facing infrastructure in-house, but this quickly runs into bandwidth issues and suffers from some problems similar to co-location.
The cloud offers some crucial advantages. For instance, Amazon's infrastructure, at least in theory, allows rapid changes in scale for reasonalby quantifiable costs for usage. Again, in theory, large data centers like Amazon's can be significantly more stable. A co-located rig attacked and destroyed (sadly still possible, especially as you open ports onto the net), can be difficult to recover from. Destroyed cloud VMs can (again, theory, practice is always YMMV) be put back up quickly.
I am in the midst of making decisions for a client about this and putting a plan in place is proving more challenging than you would think.
Re: Maddening problem
With the cloud it is easy to start and easy to pull the plug again - just pay with a CC and kill the card if the venture did not work out. And that is the End of That.
If one invested and bought a lot of kit, that capital is Gone - if the money was borrowed, the debt will remain after the kit was canned (Old gear goes for nothing at auctions).
There is some insurance aspect involved in "clouding" and "leasing".
Re: Maddening problem
I have the opportunity to put customer's stuff into our own data centres, on site and into Azure and I cannot calculate what is exactly right. Currency fluctiations, variable and changing requirements, SLA issues and so on make any choice likely to be 'wrong' in the medium term. So settng a hard limit is advisory at best, but as others have said I have seen some way too high numbers being paid for not much. But does the customer make good calculations - nope, so maybe a simple figure has benefits
10K still the trendy figure I see
Back at the tail end of the dot com era, Loudcloud, who had previously specialized in running big managed websites for third parties, decide to move downmarket by offering a "10K a month" option...just before EDS bought them :-)
Sanity at last
I'm giving this article a big +1 for injecting some sanity into the cloudgasm.
""I would agree with that," Mark Jones, a SoftLayer product innovation veep, said. "If you're spending $10,000 on public cloud [VMs] that's a pretty substantial footprint - there is a cost associated with that.""
Can't argue with that. I'll hazard a guess the cost is in the region of $10,000.
Go home with your servers, or maybe go for a PaaS like Google AppEngine. It scales automatically. Failovers are automatic too. Data is automatically replicated across two or three datacenters. I do not know how cheaper or more expensive it would be compared to Rackspace (well, it's hard to be more expensive than Rackspace) but at least you get more than dumb VMs. And you get none of the "maddening problems" mentioned above.
(disclosure: I work at Google)
(disclosure II: independently of that, PaaS is the future, the rest is legacy)
Well, yes a and NO.
With PaaS you have to have extra care not to go into a full vendor lockdown. And many times if not all, that is impossible.
So many options, so few brains
As it was stated before - finally a bit of "hold your horses" in the cloud department. Recently I was talking to a prospective customer - and I told them, that basically clouds are cool as long as they work. He said their mail & backup clouds never fail - and they want MOAR CLOUD! I gently brought up some spectacular amazon & azure failures. He repeated himself. TWICE. I ended the discussion, because it had no sense.
For a company whose liquidity and core business come from the fact, that they can communicate - mostly via e-mail - and run some shitty apps - betting all on the cloud is IMHO bad.
The important question still remains - where's the balance? Where to go? How to do smart scaling while keeping your pocket happy? (and free of vendor-rapists) I can tell ya - whether it's a web shop or 3k emps mini-corp - none of them can survive without their data and money making services.