Cloud // Infrastructure as a Service
Commentary
3/12/2012
02:28 PM
Art Wittmann
Art Wittmann
Commentary
Connect Directly
LinkedIn
Twitter
RSS
E-Mail
50%
50%

When To Pick Up Cloud As A Tool

If you use infrastructure-as-a-service like Zynga and DreamWorks now do, it can be a good deal.

For a few weeks now, I've been offering some thoughts on public cloud pricing, going so far as to say that infrastructure-as-a-service is a bad deal--which, over time, and sold the way it is now, it is. But that doesn't mean IaaS is never useful. It does have good uses, and I'll describe some here, but it's also likely to evolve into something more interesting.

Hewlett-Packard last week began to describe the services it will offer in its own cloud, beyond the standard fare from Amazon Web Services (AWS). Some of those HP services sound good; some sound problematic. But before I dive into those, here's a quick reminder of the problem with IaaS.

Buyers of cloud-based storage and compute services are paying for something whose costs become cheaper on a logarithmic scale over time, but cloud provider prices now decline only linearly over that same time. Under those conditions, sooner or later, it becomes cheaper to own and manage the resources yourself than to buy them from a third party.

Recent evidence shows that big customers are starting to reach the same conclusion. In his InformationWeek report on cloud ROI, contributor Jonathan Feldman points out that online game maker Zynga recently moved from an 80/20 mix of cloud/in-house data center resources to a 20/80 one. Similarly, DreamWorks Studios bought from AWS about 20% of the CPU hours it needed to render animated movie Kung Fu Panda 2, but it did the rest in-house.

[ Learn more about the cloud's value proposition. See Does The Cloud Keep Pace With Moore's Law? ]

If you use IaaS like Zynga and DreamWorks now do, it can be a good deal. Zynga now uses the cloud for proof of concept. It launches a game there, learns what the resource demands will be, and then brings it back in-house when it understands the need. DreamWorks used AWS for bursting (Panda bursting?) in its production work. One can imagine that as a film comes toward deadlines, the cost of using outside resources is fairly easily justified, and there are, no doubt, parts of the rendering process that use gobs of compute power.

Feldman offered an analogy most homeowners will appreciate. He looks at IaaS in the same way he does the local United Rentals store. If you're a weekend warrior and you need a tile saw, jack hammer, or concrete chainsaw for a small project, you go rent one, even though you know that a single weekend's rental price is probably a quarter of the price to buy the tool. But if you lay tile for a living, you buy your own tile saw. The point is that for very infrequent use, the price of the rented tool almost doesn't matter. You need it when you need it, and you don't want to own it under almost any circumstance. If you have computing needs like that, the cloud is for you.

Zynga's cloud strategy removes much of the risk of launching a game because the company can delay the data center capital outlay. Developers can make changes and operators can determine just what infrastructure will be needed when the company decides to bring the game in-house. And if the game flops, it flops. So when Zynga does make a capital outlay in support of a new product, it already knows a lot about a game's chances for success.

Global CIO
Global CIOs: A Site Just For You
Visit InformationWeek's Global CIO -- our online community and information resource for CIOs operating in the global economy.
As HP describes its new cloud offerings, it seems to implicitly recognize these sorts of uses. In discussing the cloud tools it's offering to support developers, HP mentions languages like Ruby, Java, and PHP. Supporting developers in their use of these languages is an implicit recognition of the types of cloud uses Zynga and Dreamworks represent – that is, for proof of concept, for testing, and for exceptional needs, all instances where developers will be key players. On the flip side, HP says it will offer structured and unstructured database services, as well as analytics.

This last area may be a tougher sell. If HP offers a "big data analysis" cloud, the gotcha will almost certainly be in the cost of storage. Because unlike processing power, which you can more or less use like electricity (on when you want it, off when you don't), storage is different. At OC-3 network speeds, you can move only a gigabyte per minute or a bit more than a terabyte per day. There's no concept of moving data into the cloud "just in time" to do big data calculations; it takes days to weeks over most wide area networks. Big data usually accumulates and stays put, and that puts us right back into the original cloud pricing problem I've been talking about.

The cloud will find its niches, but it will be a very, very long time before "everything" runs in the public cloud.

As enterprises ramp up cloud adoption, service-level agreements play a major role in ensuring quality enterprise application performance. Follow our four-step process to ensure providers live up to their end of the deal. It's all in our Cloud SLA report. (Free registration required.)

Comment  | 
Print  | 
More Insights
Comments
Oldest First  |  Newest First  |  Threaded View
parkercloud
50%
50%
parkercloud,
User Rank: Apprentice
3/12/2012 | 11:37:00 PM
re: When To Pick Up Cloud As A Tool
Simple question, how can Public cloud be compared against Private except on a case by case basis when there is no Private Cloud design standard? The answer is it can't, but it has been proven for example by Zynga that Private cloud can be done more cost effectively
ArtWittmann
50%
50%
ArtWittmann,
User Rank: Apprentice
3/13/2012 | 1:49:19 AM
re: When To Pick Up Cloud As A Tool
Take a look at the first article linked above and you'll see what I'm talking about. At least at present, cloud services aren't keeping up with Moore's Law, so you actually can say that at some point, if things don't change, the public cloud will be more expensive, at least for raw compute resources. It's just a matter of time.

I suspect that time came for Zynga, and that had a lot to do with the timing of their switch to their private cloud architecture.
Sam Iam
50%
50%
Sam Iam,
User Rank: Apprentice
3/14/2012 | 7:55:28 AM
re: When To Pick Up Cloud As A Tool
Yes, it is true that for "big data" workloads you do not get the "elastic" benefit of the public cloud from a storage perspective, but that is over come by companies like IBM or HP being able to buy their own storage at a much lower cost than you can buy storage and storage management sw... and in much greater volumes. It is true that Walmart does not have a tremendous on-shelf advantage over the local grocery store as they both have to put appx. the same number of grapes on the shelf at any given time, but they don't pay the same price for grapes that the small store pays for grapes. Less true for AWS as they don't make their own gear, but they are not paying the same price for an x86 server as the company buying 10 a year.

Also, HW is cheap, HR and SW are expensive. The public cloud providers can automate an OS patch, for instance, and apply it to 10,000 nodes at once. They can spend the time creating terrific, specialized IT management tools for their environment which would not be cost feasible for a smaller company. The cloud providers are doing basically no OS, storage admin and little hypervisor and network admin. There are some serious automation factors to consider. Sure, you could build all of this in a private cloud, but unless you have serious volumes, it doesn't make sense to spend the millions to buy/create this sophisticated automation.

The reason the United Rentals analogy does not make sense is that there is no almost constant asset utilization by the rental company. The reason the rental of a saw is 25% of the purchase price for a weekend is that the saw, over its lifetime, spends most of the time sitting on the shelf. Not so with cloud compute assets. If you return it, those procs don't sit idle for months waiting for the next renter, they go back into the pool and are spun up very rapidly. You can rent compute resources from IBM or AWS for three-four years for less than the cost of buying the servers-storage-network through their depreciation cycle. It would cost you many, many times the cost of the saw to rent it from United Rentals through its entire depreciation cycle. There is also no applicability of automation in the analogy.

The more appropriate analogy is the electrical utility. Companies formerly created their own power for their manufacturing facilities and other needs. It was skill intensive, capital intensive and took tons of time. Then utilities came along (after electrical distribution, the internet of electrical power, became efficient). They provide three advantages over having a power plant on-site: 1) economies of scale. 2) automation 3) no need to worry about operational commodities, i.e. keeping the lights (computers) working. 1 and 2 = lower costs. 3 = More time to focus on competitive differentiation, less on operations. Now there are companies that have power plants on-site and do not use utilities, but they have huge electrical volume requirements which rival utility scale.

There is one major difference between electrical power and compute power. You don't have to audit and secure your electricity as you do with your data. Compliance and security validation is the major hurdle for public cloud.
ArtWittmann
50%
50%
ArtWittmann,
User Rank: Apprentice
3/14/2012 | 2:53:17 PM
re: When To Pick Up Cloud As A Tool
The question of the right analogy is a fair one. One reason why I think United Rentals is a reasonable analogy now is that UR doesn't have a lot of competition. UR spends a lot of capital to have resources that no one else has - maybe the tile saw is the wrong tool to look at. A cherry picker might be better. They have dozens, no one else has even one, not only because of the cost to buy, but also because of the cost to maintain. Sounds a lot like your argument about scales of economy for IBM and HP.

I think that what you say about scales of economy translating into cheaper systems for cloud providers to own and IT orgs to use could be true come day, but the market isn't there yet. Vendors are not yet in the habit of passing back to customers savings accrued from advances in technology.

Part of the problem may be the pricing model itself. As you say, hardware is cheap and gets cheaper, other factors don't. A pricing model that reflected this might have better sustainability over the long term. But if you sell by the byte and by the cycle, either you keep dropping prices, or you invite customers to figure out when they can run the same resources more cheaply themselves.

The electric utility model is interesting, but not right. For one thing, it's a regulated industry, and California has seen just how much trouble you can get into when you get the dance between provider and regulator wrong. I have no idea what would happen to electricity prices if you completely removed regulations, but I'm pretty sure I don't want to find out. Once you end up in a market subject to supply and demand, someone's demand could dry up your supply, and that's not something IT shops or CEOs are ready for.
Sam Iam
50%
50%
Sam Iam,
User Rank: Apprentice
3/18/2012 | 5:32:43 AM
re: When To Pick Up Cloud As A Tool
But companies are not buying cherry pickers (highly specialized workloads) from IBM SmartCloud, Azure, AWS. They are buying hammers (commoditized platform workloads), such as vanilla MS Server instances or vanilla RHEL instances. UR has scale advantages, but people use UR not due to pure scale advantages. They use them, as you mention, for rare needs. Utilities, like cloud services, sell a commodity service with a pure scale advantage.

"But if you sell by the byte and by the cycle, either you keep dropping prices, or you invite customers to figure out when they can run the same resources more cheaply themselves."

I don't know why this would be a problem for the "cloud" services industry, unless there is price fixing going on between the cloud providers. If someone is making a huge margin by not passing on Moore's Law savings to customers, another company will undercut them. I don't know why the market rules would not apply. If IBM or HP do not lower their server prices in line with advances, the other will start to gain share with lower prices. I don't know why this would be different for IBM and HP in the cloud market.

The benefits of the utility, or cooperative, model are not negated by regulation. Controlled monopolies allow utilities to build up scale that they would not be able to have in a completely open market, but the level of scale is the only thing that would change in an open market system. Without regulation, power cooperatives still would have the benefit of scale beyond what most individual businesses could achieve on their own, buying power beyond what individual businesses could achieve on their own, automation of processes through investment in tools which are ROI justified by scale, etc.

"Once you end up in a market subject to supply and demand, someone's demand could dry up your supply, and that's not something IT shops or CEOs are ready for."

CIOs are already in this world. Look at memory prices and scarcity over the last five years, sky rocketing due to demand for virtualized workloads. I don't know why the world changes when these servers are in a larger cloud provider data center instead of a smaller data center. If demand goes to the moon, e.g. large company x decides to buy 10,000 of a particular server, other companies are going to wait or pay more for those servers. The same would be true if company x buys the equivalent of 10,000 servers from a cloud provider. The others will have to wait until capacity is added or the prices will go up.
Multicloud Infrastructure & Application Management
Multicloud Infrastructure & Application Management
Enterprise cloud adoption has evolved to the point where hybrid public/private cloud designs and use of multiple providers is common. Who among us has mastered provisioning resources in different clouds; allocating the right resources to each application; assigning applications to the "best" cloud provider based on performance or reliability requirements.
Register for InformationWeek Newsletters
White Papers
Current Issue
InformationWeek Tech Digest - July 22, 2014
Sophisticated attacks demand real-time risk management and continuous monitoring. Here's how federal agencies are meeting that challenge.
Flash Poll
Video
Slideshows
Twitter Feed
InformationWeek Radio
Live Streaming Video
Everything You've Been Told About Mobility Is Wrong
Attend this video symposium with Sean Wisdom, Global Director of Mobility Solutions, and learn about how you can harness powerful new products to mobilize your business potential.