Archive for June, 2011

The Cloud SPI Model Part 1: Software As A Service, What You Really Want From the Cloud

For the most part cloud products have sorted themselves into three classes: Software-As-A-Service (SaaS), Platform-As-A-Service (PaaS) and Infrastructure-As-A-Service (IaaS). These three offerings are now being referred to as the SPI Model of Cloud services. So what are the differences, what should you get and who offers what?

As we started deliberating the thinking around SPI, it became apparent that this would be a very long post, so it’s broken into three parts, one for each offering.

It’s important to realize that SaaS, PaaS and IaaS are not mutually exclusive – they build on each other. SaaS is only possible because under the hood there is platform and infrastructure that SaaS depends on. The platform and infrastructure may or may not be available for sale as PaaS and IaaS offerings, but they are there.

All Cloud offerings are ultimately SaaS, it’s just a question of how much you build yourself. With SaaS, you’re building nothing – the application is finished, you just configure it the way you want to use it. Some of these applications have been around longer than the term “Cloud”. Products like have been sold under the Application Service Providers (ASP) moniker for a long time. As the larger concept of the Cloud came to the forefront, they jumped onto the bandwagon, but it didn’t really change the product.

Changing the moniker didn’t change the application, and the users couldn’t care less. Users don’t care about Cloud or any other technology – they care about getting their work done. How you deliver the tools to them that let them get that work done isn’t important to them. If that tool happens to be software, and that software is delivered via Cloud technology, and it is effective for the user, then the user will like Cloud. But that begs the question, what makes a given technology (like software) a Cloud offering?

The best definition we’ve found for determining whether a given technology is “Cloud worthy” comes from Dave Neilsen of Platform D. The acronym is OSSM (“awesome”): On-Demand, Self-Service, Scalable and Measured. On-Demand means the product is available whenever you want in whatever quantities you want. Self-Service means that you can order the product without needing any support from anyone else (especially the vendor selling the service). Scalable means you can order as much or as little as you want, it works the same regardless. And finally, Measured means that the product is measured in reasonable increments so that you know exactly what you’ve bought and how much it cost. You’ll pay only for what you use.

Using the OSSM standard, it’s easy to see that qualifies as a Cloud SaaS offering – you are able to order it yourself without any assistance and get to work immediately. You can buy as many seats as you want, enter as many leads as you want, run as many reports as you want. And you only pay for what you use.

Other SaaS offerings including Google Docs and Microsoft Office 365. In fact, mail in the Cloud is arguably the definitive SaaS product. Why go to the expense and effort of operating your own mail servers when you get everything you want from a SaaS offering, typically at a fixed rate per mailbox?

There are a couple of challenges with SaaS. The first is security. For the most part, security for SaaS applications is sufficient, even if it’s nothing more than SSL. But if the security features of a SaaS application don’t comply with your organization’s requirements, there’s often no recourse. The security the application has is the security the application has, take it or leave it.

A more complex issue is regulatory compliance. This is a huge subject unto itself, depending on the industry you’re in, the countries you’re working in and the work you are doing. There are whole web sites dedicated to the topic of regulatory compliance like the Compliance Authority that also talk specifically about compliance and Cloud. Smart SaaS vendors are helping their customers get through regulatory compliance concerns with great documentation and even features – for example, Microsoft’s Office 365 allows mailboxes to be stored outside the cloud so that a customer can fulfill compliance requirements for mail from a particular country to be stored in that country. There’s also sophisticated mail retention policies and auditing.

Beyond these issues, there’s almost no downside to SaaS. And there is a huge upside –  all the things you don’t have to do. You don’t own the hardware, the operating system or even the software you’re running. Someone else is responsible for all of that, you just utilize the product as you see fit. From an ROI perspective, SaaS is a no brainer.

Ultimately, SaaS is what we want from Cloud – the maximum potential with the minimum of ownership and responsibility. But what if the application you need doesn’t exist in the Cloud already? That’s where Platform-As-A-Service (PaaS) comes into play. It’s the next layer down the SPI Model stack, and is the subject of the next blog post.

You sold me on the Cloud. Now what?

After tons of reading, attending numerous events and speaking to peers in your industry you became convinced: The cloud is where you need to take your company going forward. Now what? Like any major project you’ve taken on, going to the cloud requires a strategy and a roadmap.

Strategy has nothing to do with cloud per se – cloud is simply one potential solution to a particular strategy. So what’s your strategy? Are you trying to reduce infrastructure costs? Enable more work-from-home employees? Improve your scalability to deal with massive surges in sales? Handle a more mobile, more connected, more bring-your-own-hardware work force? Perhaps your strategy is a combination of all these things (and more), but it needs to be articulated and then you need to prove that the cloud offerings you are considering are able to fulfill that larger business strategy. Cloud is often only one peice of the puzzle.

The roadmap you build is your path of execution. Perhaps you’ll start with a straightforward milestone of migrating email to the cloud using any number of Software As A Service offerings. These are the low-hanging fruit opportunities that are currently bundled under cloud services. But there are more challenging milestones ahead… Infrastructure As A Service and Platform As A Service offerings require a lot more consideration, a lot more effort and carry a lot more risk-reward potential. Rewriting a web site to take advantage of the elasticity of the cloud can be a very risky effort – you’re taking a web site that currently works and potentially breaking it. But the ability of the cloud to absorb a massive influx in traffic can pay back in revenue all that risk and more.

The amount of hype around the cloud makes it possible to believe that there’s a magic cloud lamp out there somewhere – that all you need to do is rub it and your company is in the cloud. The reality is much more painful. Inflexible applications, questionable security and technology silos make your flight the cloud more than a little turbulent. Some bits of cloud migration may look easy, but most are not.

As you work your way through your organization evaluating how cloud can bring your strategies to fruition, you’re actually going through a form of business process re-engineering. Don’t Panic! This is a natural side effect of any major IT project. You cannot help but scrunitize what your organzation currently does and ask the question “Does this make sense?” and “Could we do it better?” Moving to the cloud is not unique in this effect, nor is it separate from it.

Ultimately, you’ll discover that virtually all cloud implementations are not “pure” per se. There are always some elements of an application or service that still reside on-premise. For the foreseeable future, all significant cloud implementation are hybrid implementations. Don’t feel bad about it, it’s the way things need to be, perhaps indefinitely.

You’re sold on the cloud, so congratulations! Strategize, plan and move carefully. We’ll see you out there.