Making a Cloud Play
Wayne Beekman is maneuvering his business to catch the cloud. Should you be doing that too?
When it comes to cloud computing, Wayne Beekman is ahead of the curve.
Beekman is principal at Information Concepts Inc., a custom-solution development shop based near Washington, D.C., in Herndon, Va. In a time when most established partners are eyeing cloud computing warily -- and from a distance -- Beekman has gotten up close to the cloud-computing offerings from all three major vendors: Google Inc. with the Google App Engine; Amazon.com Inc. with its EC2 and other platforms; and Microsoft with its Azure platform, which is set to launch in November.
While he's taken an agnostic approach in his evaluations and recommendations to customers, his perspective is uniquely informed by his Microsoft experience. Information Concepts was founded in 1982 and has been a Gold Certified Partner since Microsoft created the role in the mid-1990s. Meanwhile, Beekman has participated as a member of the .NET Partner Advisory Committee since 1999. In September, Redmond Channel Partner's Scott Bekker got on the phone with Beekman to talk about his experience setting up a cloud-computing practice a year ago. He also got Beekman's perspective on how cloud computing will play out across the industry as Azure goes live commercially and as the concepts of cloud computing become more mainstream.
How did you get interested in cloud computing?
A year and a half ago, I saw an ex-Microsoft employee blog post that said, 'Hey, look what Google's doing.' It was the announcement of the Google App Engine. So I sat there for an hour looking at their videos. I was just fascinated by the concept -- to not have to provision or procure any hardware and have it instantly accessible without having to deploy.
Google was having a show in San Francisco. There were 4,000 attendees -- it was sold out. There was a vibe in the room. [Before that], I thought, these are just some developers from some startups. But as I walked around the room, I saw a number of Fortune 500 name badges.
I came back, and I immediately hired a woman to go out and talk to all of our existing commercial customers. That was a year and a half ago.
What's happened since then?
A year and a half later, I've gotten better meetings at higher levels in organizations than I ever had since I started the company in 1982. Traditionally we'd meet with a director of an application development group. Our meetings over the last year have been more at the CIO level or the CFO level.
In the first year of this, I felt like I was doing a lot of education. I came up with the stacks [a set of diagrams], which helped people visualize what I was talking about. I was able to talk about specific requirements based on the stacks. Basically, I had to build my own tools. That was year No. 1, and we've got a little business around that.
In the last few months, I've had more meetings where executives at organizations are saying, 'I'm committed to moving my stuff to the cloud. I understand it.'
They understand the difference between infrastructure as a service and development as a service. Right now, I'd say they're more interested in infrastructure as a service.
How have the customer questions and objections changed?
What I've heard in the last couple of months are a different set of objections. Originally, if you asked what customers' top 10 questions were, the top three were security, security and security. The next three were SLA [service level agreement], SLA and SLA. The other four were just that they didn't understand it. They also asked, 'How do I know this is going to be around?'
Now the questions are much more sophisticated. One CIO said, 'I'm committed to this, but I already have hardware that I'm paying depreciation costs on.' Another CIO said, 'I'm committed to doing this. The problem is, how do I start? I have a purchase order for a new piece of hardware. I have to rebuild my operational procedures. I've got to get there. I'm just not sure how I get over these organizational hurdles to get there.'
On the other hand, I had a session with an organization, and, after an hour and a half of understanding what we were talking about, they actually identified a server in a cage on the west coast that had to be offsite. They could actually shut down the server and save $10,000. At the end of the meeting, one of them says, 'Oh yeah, I'll sign on to Amazon tonight and take care of that.'
So you've saved a customer $10,000. What's in it for you?
[Laughs]. Well, that's the tricky part. You can probably cherry-pick a couple of things to establish yourself as a trusted advisor. The bottom line is that as solutions providers, .NET isn't new. There's a lot of competence in .NET development shops. One of the reasons we're investing in our cloud-computing practice is that it provides a different way to bring value to our customers. We're going to architect the solution differently for you, which is going to save you a ton of money. We're coming up to almost 10 years of .NET. Sure, you can be a great SharePoint provider, but at some point, services become commoditized.
Historically speaking, we're primarily on the development side. We always work with the operational groups to make sure that what we build [can get implemented and supported]. If it's a small customer, we're doing everything. I've had more discussions lately on the infrastructure side, especially with Amazon Web Services.
The value proposition that I'm focusing on is in the area of disaster recovery. We bring value to traditional infrastructure folks that we traditionally didn't bring value to. As a side note, Amazon Web Services just announced their partner program, and we were put in as one of their solution providers. Disaster-recovery scenarios are one of the things that Amazon is emphasizing.
Would we love to just build brand-new apps from scratch? Yes. But in today's economic world, our customers are looking to save costs. They're saying, 'My data centers are full. I can't take it any more. I need to start to move out.' They've all gone through virtualization. They've maxed out what they can do in terms of virtualization. As a solutions provider, with Microsoft expertise, we can provide value in that space.
How has the cloud-computing practice worked out for Information Concepts?
When I started it, I thought within eight weeks I'd just put it on the shelf and say, 'That was a fun experiment.' I had no idea [about] the level of traction. After the first eight weeks, I realized that this thing has legs.
In terms of revenue, we received some traditional project work from these discussions. Customers say, 'Hey, how about this type of work?' We did get incremental revenue into the organization in the first year.
If you fast-forward to what we accomplished in the past year:
- We reinforced our relationship with Microsoft.
- We established ourselves with the developer evangelist teams at Google for Google App Engine, and we have really good access if our customers have questions.
- We joined Amazon's partner program in the very first round of companies.
As far as where it fits into our existing business, our traditional practice area is vertical. The cloud-computing practice is going to have to intertwine itself across all of those areas. In the last year, some of our senior developers have drilled down on some of the services and how to architect solutions. About a year ago, we moved our day-to-day backup to the cloud. I figured I couldn't go to customers if we weren't using this ourselves.
We've built some prototypes. We've moved some of our internal systems to use cloud-based development tools. We've taken our existing issue-tracking systems and built Web-based front-ends to them.
But for current projects, we have to service our customers for what [they] signed up for. As new projects come in, we're going to start to integrate these services. It's evolutionary to the process.
I think we've reached our goals. It was a tough economic year for a lot of our customers.
We've built out enough from a proof-of-concept and relationship standpoint that we're ready to take on the projects now. We want to be the go-to company for cloud-computing development in the Washington, D.C., area.
How will you decide which platform to recommend to a customer -- Google, Amazon or Microsoft -- once Azure is available?
We're agnostic when it comes to the cloud platforms. It depends on the solution the customer is looking for. If I come into a shop that's doing .NET development and has great experience with SQL, Azure is going to be the fit. If I [meet] a customer that isn't doing net new development, but they want to forklift an application into the cloud, then we're going to talk about how we can use Amazon EC2. If I walk into a shop and they're all about Java or Python -- an open source type of shop -- then we're going to talk about how Google can enhance what they're doing. We're not going to change a Java shop into a .NET shop.
The question isn't should I go with Google, Amazon or Microsoft. The decision is: When do I go with cloud-based resources? By providing strategic assessments and support to their planning, we can help customers navigate that choice.
What do you think other Microsoft partners should do with cloud computing?
As a .NET development partner, it's important for us to not only know what the platform is today, but [also to know] where it's going. Azure is just another piece of the platform. That doesn't mean the customers are going to adopt it, or we're going to force it. It's another tool in the bag.
You need to be able to understand what it is. What are kids doing in college today? They don't have Windows servers sitting in their dorm rooms to build really cool applications -- but they do have access to the Google App Engine. What's going to happen in three to five years when those kids start coming into the workforce? You're going to come in with a proposal for $100,000 of infrastructure costs, and there's going to be some kid fresh out of college saying, 'I think I can do that for nothing.' You have to understand the tradeoffs. There might be a reason to spend $100,000 -- there might not be.
I'll give you an example of an application a couple of years ago where it would have made sense to bring cloud computing into a hybrid solution.
We built an application with 100 users internally and 35,000 external users who come in at various points of the year. The customer needed to build an on-premises server farm to support peak load when those 35,000 users decide to sign on. Most of the time there were 100 internal users. If we were to architect that solution today, we would strongly look at having the presentation and transaction portions on Azure. Or we would explore with the customer whether the cloud could handle the transaction volume for those external users. It may be the 100 internal users we run in an internal environment.
With cloud computing, there should be no reason to procure and provision hardware for peak load if it's very spiky.
We're looking at hybrid solutions, not throwing the big switch. I don't want anyone saying after we go to a meeting, 'Oh, Beekman said we've got to get rid of our data centers.'
I understand you'll take an agnostic approach, but what's your take on Azure, given your Microsoft focus over the years?
Azure is fun to talk about. It's great because people get it. It's a natural extension to a .NET developer's toolkit. That's what I love about it. You can take .NET code and be able to extend it to the cloud.
Microsoft isn't creating new database access methods in order to do what we do in the Microsoft space. We like joins, we like [SQL] statements. We have to change the way we architect things less by using Azure. It's more of a natural extension. We'll see in November what really happens in terms of adoption.
This stuff is just so exciting, but I don't think Azure has got the same level of attention. Folks don't want to talk about Azure as much as they want to talk about Amazon and Google. Because Azure is such a natural extension to .NET, it's not as exciting to talk about.
In customer conversations about the cloud, what's surprised you as far as what they're interested in or how they want to take their applications to the cloud?
Until recently I didn't understand the value of infrastructure as a service for existing workloads until customers started bringing up disaster recovery scenarios. I think disaster recovery is a great use case. Why do you need to pay for a bunch of machines in a cage under a mountain somewhere?
Build up images, keep a couple idling. That really came from customer input.
The other thing that came from customers [was]: 'I'm a transaction guy. I'm a database guy.' We thought originally that this is extending existing databases. We went to organizations that were broadcasters and publishers and other media companies. They don't want to consume the bandwidth. Here's an example with Amazon's S3 cloud offering for storage. Putting the storage out in the cloud and having the applications deliver it to customers that way is really exciting to them. I didn't understand the value of using storage in the cloud that way until customers brought it to us.
Many of the use cases wouldn't be there today if it weren't for customer discussions.
Why did you believe that Information Concepts needed to take a hard look at cloud computing?
You really need to add value to your customer discussions. The slogan of our company is, 'Solutions that fit.' Since 1982, if we hadn't brought something new to our customers, we wouldn't be able to stay a trusted advisor or a valued resource for them. The reality is that, since then, the process hasn't changed. You've got to listen to customer requirements, you've got to design a solution that's going to meet those requirements, you've got to build and test it and you've got to deploy it. Cloud computing has been an investment, and, looking back over the last year and a half, I think it's on track. There are some incredibly exciting projects that we've been talking to customers about.