We have been doing some work with with Amazon’s Elastic Computing Cloud (EC2) which allows us to create virtual machines in the cloud in a few seconds. These are great for hosting websites, and what’s cool about them is that if you get Slashdotted or experience a similar unexpected spike in traffic you can create new hosts immediately. Recently Amazon added a new service called Elastic Load Balancing (ELB) which can distribute load across hosts. We’ve been looking at this for some of our recent development and infrastructure projects.

I just read this description of how ELB works by Shlomo Swidler from his Cloud Developer Tips blog. It’s a great reference.

You pay for ELB by usage just like everything else at AWS. From Amazon: “You are charged at $0.025 per hour for each Elastic Load Balancer, plus $0.008 per GB of data transferred through an Elastic Load Balancer.” For reference, on a deployment project in 2008 our Engineering team used a Cisco load balancer which I imagine cost a few thousand bucks.

Cost isn’t the only advantage. These can be created and destroyed quickly and remotely, allowing us to work more efficiently and spend less time visiting data centers in the middle of nowhere. This leads to improved quality of service for our clients as we can spend more time consulting on future technology growth plans and less time troubleshooting servers in cold, loud data centers.

This blog post brought to you by the iced coffee I am enjoying in the comfort and quiet of my office while deploying virtual machines!

Tom Mills from Google and Sean Finnegan from Mozy will be giving an in-depth review of their offerings and how they fit into an agile, post-recession office technology strategy.

I’ll be giving a brief overview of how I think we arrived at this point in IT and what it means for creative, innovative firms that are trying to do more with less.

In an effort to get my thoughts together and get some feedback, I’m using this blog post as a draft for the event.

Let me start off by giving a brief overview of our services, and then a little history about the evolution of our offerings:

We provide a number of technical services for our clients in the areas of infrastructure, application development, and industry-focused workflow consulting. As this is New York, we work with a number of creative firms; media, architecture, publishing, and design companies, as well as some key clients in the financial sector. We strive for long-term relationships with our clients, many of whom we’ve worked with for close to a decade. We have installed and managed hundreds of servers, network devices and application suites, but more recently we’ve been focusing on helping our clients select, migrate to, integrate, and manage Cloud-based services.

Since the 1990s and the introduction of pervasive bandwidth, we’ve gone through a number of permutations of the remote server/client model, and much has been written about the benefits and the irony of the shift back to the mainframe/thin client structure of the 1960s. Now everyone is talking about the future of ‘The Cloud‘; a vast array of computing resources, abstracted and presented as a single source to the consumer.

At the turn of the century, we found most small to mid-sized businesses with a pure Local Area Network (LAN), typically comprised of in-house mail – most likely Exchange – and a few other local services: file, print, etc.  A lot of these firms had an internal IT staff or a dedicated consultant to manage their servers, tape backup, networks, and desktops. Only a few were pushing the envelope by leveraging Application Service Providers (ASPs) to deliver back office services.

The risks with this situation were obvious. These systems mostly depended on a single Internet connection, a single building, and a single individual, prone to career changes and untimely vacations.  Remote access to these in-house services was expensive to do right and applications rarely worked as well remotely as they did in the office.

Over the next five years, we saw a gradual shift towards ‘Hosted Applications’. This typically came in the form of a service provider taking a LAN-based solution like Exchange or SharePoint out of the office and putting it in a data center. In conjunction with this change, we saw the IT services industry begin to shift its focus from in-house IT, or consultants, to managed services – companies providing regular systems management remotely.

There were some benefits to this offering: critical applications were not dependent on intermittent Internet connections or over-heated server rooms. Flaky consultants were traded for predictable management services and cost became as regular as the electric bill.

But there were still problems. We had the same old model of doing things, only it was moved out of the business’s office and into the provider’s.  Services that were built for an onsite installation and LAN speeds were shifted to a remote location – not always producing the best results. Access to applications designed for the LAN was sometimes unacceptable because of bandwidth and latency. In a similarly narrow view of the problem, Managed Service companies focused on monitoring systems and patching software, maintaining the status quo, without looking at the big picture, or driving the business forward.

Now the next generation of IT services is coming along and delivering on the promise of on-demand, scalable solutions. These services are web-native, built for the Cloud and multi-tenant environments.

As services like Google Apps and Mozy were built for the web – not re-purposed LAN applications – they deliver exceptional performance and remain very flexible. Control Group has designed our support and project services in a similar way. Our services are built to function efficiently remotely – scaling up when our clients need it, and going away when they don’t – and also to be flexible and innovative, driving business forward rather than maintaining the status quo.

Using the cloud paradigm, we act as a single source of technology for our clients. We help them run more efficient, profitable businesses by weaving an ever growing selection of web-based services, traditional IT, and industry expertise together, to provide an flexible, competitive business platform.

Your Data Here

I was having a conversation recently with one of our consultants, David Rocamora, as our team broke down the contents of a start-up’s data center, when I came to the realization that we may have built our last data center.

Now we don’t really build data centers, but we have racked a lot of servers, storage and network gear around the world in tier 1 data centers for our clients. With the change in the economy and the maturity of several cloud services, the data center that we know and love looks like it’s going the way of the wood pulp newspaper.

Certainly more data centers than ever are being built: Google, Microsoft, Apple, and Amazon are soaking up gigawatts (petawatts?) of power like never before. But the days of dressing down in jeans and a sweatshirt and going out to New Jersey or Colorado to rack servers in an earsplitting, freezing cold warehouse of caged servers and blinking lights seems to be drawing to a close.

Some numbers to consider:

This ‘scaled down’ dot com we were consolidating, had spent about $500,000 on a few racks of some amazing equipment (Sun, Check Point, etc) only to find out 4 months later they didn’t need it. Pennies on the dollar. The contract for the floor space, power, and bandwidth goes on for another 8 months and I bet you could buy a modest BMW for what it’s costing them.

Now a similar sized start-up we just started working with on a really innovative interactive image platform, is using the Amazon cloud and RightScale and is spending about $50,000 a year on cloud services. No capital outlay.

Flexibility?

The cloud is infinitely more flexible, we can put servers in Europe in a matter of minutes, set up high availability zones in different regions around the country, and if they start to get swamped with business like we think they will, we’ll be able to turn up as many servers as they need in a few minutes time.

What if the dot com of 4 months ago took off? Order servers, spend capital. Put in a request for more bandwidth, more cage space. Days, maybe weeks go by. Then get out the jeans and sweatshirt and head over to the data center. Earplugs. Man, those servers are loud.

You get the picture. But this is happening so fast it’s amazing. Six months ago when the dot com was building its data center, the Amazon cloud was still in beta, with no SLA, and it wasn’t an option for a serious start up. Today, building a data center isn’t an option for a serious start-up.

Now we have availability zones, provisioning and monitoring tools, the ability to drop terabytes of data into the cloud — shipped through FedEx! But the real promise is the rich API and the spirit of community innovation. Companies like RightScale are finding a niche in the cloud, developing something really valuable, and then selling it as a simple service that makes our lives so much easier.

It’s exciting to see this happen so fast.  To avoid being crushed by this wave, as an IT team, you need to really stay on top of it. IT in our part of the ecosystem is becoming more the art of selecting, deploying, integrating, managing and supporting cloud based services, and much less the craft of building serious web infrastructure.

It’s a little sad for the hardware geek in all of us, saying goodbye to the roar of the servers, putting down the Velcro ties and picking up some slick provisioning and automation scripts. But I think we could get used to deploying 50 servers in a few keystrokes from a quiet, comfortable seat in the office.