Software-Defined Data Centers Might Cost Companies More Than They Save

storagedude writes "As more and more companies move to virtualized, or software-defined, data centers, cost savings might not be one of the benefits. Sure, utilization rates might go up as resources are pooled, but if the end result is that IT resources become easier for end users to access and provision, they might end up using more resources, not less. That's the view of Peder Ulander of Citrix, who cites the Jevons Paradox, a 150-year-old economic theory that arose from an observation about the relationship between coal efficiency and consumption. Making a resource easier to use leads to greater consumption, not less, says Ulander. As users can do more for themselves and don't have to wait for IT, they do more, so more gets used. The real gain, then, might be that more gets accomplished as IT becomes less of a bottleneck. It won't mean cost savings, but it could mean higher revenues."

Comments

IT the bottleneck?

[cosm]

GTFO.

Re:IT the bottleneck?

[Zaelath]

As time marches on, people are becoming more IT literate

Hahahaha

http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect

Re:IT the bottleneck?

[plover]

Sure, IT's the bottleneck. Why? Because users who roll their own solution without understanding what they're creating will create a fragile business model.

Andy Accountant decides to do General Ledger on Quickbooks, while Polly Payables decides to do billing on the bank's web server. How does one update the other? It starts out as a manual process. But let's say Polly is clever and signs up for IFTTT.com to automate the integration. She also hands the task of entering the bills over to Carlos Clerical. Later, when Polly is on vacation, Andy downloads an upgrade to Quickbooks - but IFTTT was set up only to modify the original Quickbooks. Now Polly's billing doesn't work, and Carlos has no idea what's going on. Polly is the only support person, but she's on vacation. Andy only knows about the manual processes, so he can't help Carlos. So the bills don't get paid.

And the IT guy only knows about the PCs, the printers, the network, and the file server. He doesn't know about the apps, because the users got tired of waiting for him and rolled their own.

Repeat this scene for each and every system, service, and person involved with computers in the organization. It starts out easy and fast, but the dependencies quickly crust over every activity the company performs. Support becomes a nightmare, and changes go from "difficult" to "impossible".

If the IT guy put the pieces together, he (should) document the connections, provide troubleshooting knowledge, and at least know who to call for support. At least that's the theory.

Re:IT the bottleneck?

[JanneM]

If you had a response time of a week for issues, and you had to work enough unpaid overtime that you left rather than facing an intolerable work situation then quite obviously you were not able to handle it with the staffing at hand.

Re:IT the bottleneck?

[hawguy]

A car analogy:

This is the same type of BS that the city I live in uses so they don't have to build or expand any roads, "Well, traffic will just be as bad even after we get the highway built, so why bother?"

It's true. Once they build a new highway into town, people will build houses farther away on the other end of the expanded highway, so the new highway just fuels more suburban sprawl, so it causes more congestion inside the city and for drivers closer to the city where it may not be possible to build more roads at all. Or, in areas without a major population center, it can encourage job centers to spring up along the highway, which is difficult to serve with cost effective transit as people are forced to commute farther and farther to get to their jobs.

By not building the roads, they implicitly encourage more high density, transit friendly development closer to the city.

In general, a new highway is a temporary (and expensive) solution to a traffic problem.

Re:IT the bottleneck?

[bored]

Yah, and that is why the "cloud" providers are less expensive. Do you really think there is a 7 figure EMC sitting behind an amazon storage node?

No! see apples to oranges again. For some reason its ok, for the cloud provider to run on cheezy hardware missing most of the "enterprise" features, but its not OK for random company to buy similar hardware.

Re:IT the bottleneck?

[theshowmecanuck]

We allowed a senior developer to manage his own AWS EC2 instance for development of a server. Then we noticed a few hundred gigs of data were moving through the server when he wasn't around. We shut it down and audited it. There were ports and security vulnerabilities exposed that just shouldn't have been, because he had set it up to be easy for development etc etc etc. The IT bottleneck was removed. So was a good chunk of money from the company paying for some people at a number of Chinese IP addresses to move data through our servers (that costs on cloud services ya know). Just because people know how to make things work, doesn't mean they know how to do so safely. Nor does it mean they have the inclination to learn to do it safely either. I don't like data Nazis any more than the rest. But they serve a purpose that no-one else is willing to take on. Let the average computer users control things, and your company will be fucked.

Re:IT the bottleneck?

[hawguy]

Yah, and that is why the "cloud" providers are less expensive. Do you really think there is a 7 figure EMC sitting behind an amazon storage node?

No! see apples to oranges again. For some reason its ok, for the cloud provider to run on cheezy hardware missing most of the "enterprise" features, but its not OK for random company to buy similar hardware.

Companies want to see the big Netapp or EMC name on the array so they can trust that the manufacturer knows what they are doing enough that their data is safe. Amazon and Google can get away with using cheap commodity hardware because they *are* the big name, and people trust that they can keep their data safe, so they don't need to turn around and buy hardware from the big storage vendors.

Are cloud providers really much cheaper? An entry level Netapp FAS2240 with 12TB of disk costs around $16K

Amazon charges $0.095/GB/month, or $1140/month for 12TB. So after 14 months on Amazon, you could have bought a local array.

You still have to back up (or replicate) the data from the local array, so that's not a true apples-to-apples comparison (assuming that you trust S3 enough that you don't keep your own backup of the data). a 12TB array is pretty small so you don't get much economy of scale, so once you get into the larger arrays with 100's of TB, I think the numbers swing farther away from S3 for corporate storage.

Re:IT the bottleneck?

[uncqual]

But, keeping three copies of the data on cheap hardware, one of which is hundreds of miles away and having a couple other data centers to which the data migrates in seconds and minutes is within the scope of a cloud provider -- just business as usual (the exact number of data centers and copies is irrelevant as they depend on this years stats for the low cost hardware - it's all statistics).

A business whose business isn't to maintain ten(s) of data centers and manage them for redundancy may not be willing (nor, probably, should they) to pay for that level of redundancy just for their own ten terabytes of important data (their business is making innovative widgets efficiently, not managing geographically distributed data centers, each with a connection to at least two independent power sources plus backup generators).

If a midsized business making drywall needed another car to transport a sale person, would they build an auto plant to build that car? No, they would lease the car from a business whose business was leasing cars (and providing replacement cars and maintenance) and who, in turn, bought them from a specialist in designing and making cars (Toyota for example).

Re:IT the bottleneck?

[Anonymous Coward]

I can give you a counter example. We had a Cloud Ops team that had the task to exactly prevent this stuff you described.
Great. Except, they didn't even know how to set an EC2 instance with EBS. Also they couldn't provide the EC2 instance types that were needed.
So in the end, we just worked around them. Instead of taking *days* to explain them what we needed, we had our EC2 instance running in 5min exactly how we needed it.

Not a 1:1 ratio

[Tony Isaac]

Virtualization makes it easier to stand up a new "server." True.
This simplicity will lead to using more "servers." Granted.
But those virtual servers require far less hardware than the old physical servers. Many of these virtual servers are used only a small percentage of the time. Depending on the load, 10, 20, or even more servers can run on one physical piece of hardware.

So even if we use, say, five times more "servers" with virtualization, we will be using fewer physical units--fewer "resources."

In short, the math is not so simple.

Jevon's Paradox

[Okian Warrior]

Jevon's paradox is valid, but only under specific economic assumptions.

It's only true so long as there is more demand for the resource, and it's only a problem when the resource has a cost attached. Essentially, it's true in a "scarcity" economy, but not true under "post scarcity".

We've achieved "post scarcity" for several resources already; for example, phone calls and computer time.

Phone calls used to be expensive and billed by the minute, but nowadays it's virtually free. Similarly, computer time used to be metered and charged - in college, the CPU time for each program run was deducted from your account. Nowadays people can have as much un-metered computer time as they want.

CPU time and phone service aren't literally free, but the cost is so small as to be negligible.

Despite this, we do not see infinite consumption. People have a certain level of need for a resource, and when that need is met they stop consuming more. Coupled with a declining population, there is no reason to expect infinite consumption.

Your company may be using more resources than it needs... but so what? Computer resources are remarkably cheap - so cheap, in fact, that it may be more effective to ignore the problem. Optimize the biggest expenses first: if that turns out to be IT resources, then take a closer look. Otherwise, just ignore it.

(For another example of post-scarcity, consider the Chinese "dollar stores" that have cropped up. The cost of goods is so small that the time and expense of price tags makes a big difference. This is almost post-scarcity of tangible goods.)