The Amazon S3 outage that took down about one-third of the Internet left many CIOs wondering about their own environments — and where the weak link lies. Meanwhile, in the words of Marc Andreessen, “Software is eating the world.” Businesses run on applications — email, CRM, ERP, some customer-facing, many mission-critical.
Clients tend to add bandwidth on an ad hoc basis as they pick up SaaS applications, like Office 365, Marketo or Salesforce. Soon the environment is a hodgepodge that is largely unmanageable.
At the same time, 20-year-old networking technology MPLS is breaking. Network admins never really embraced MPLS because it was too complex. Now, a majority of the traffic is off-net (on the Internet) via SaaS, AWS, Azure and more, so the wide-area network needs to be re-structured.
What am I getting at here?
In tech, we tend to think that the technology is the product. But when it comes to cloud, technologies need to be implemented as solutions with a solid business strategy. This isn’t a product replacement situation.
When CIOs are making cloud transition plans, outages are always in the back of their minds. But beyond redundancy, there are concerns about data integrity, data backup, compliance and security. They are also aware of the risks created by sharing virtualized infrastructure with hundreds of other companies. It’s a very different world from having a traditional data-center footprint where no other company could potentially bring down the servers that are also hosting their IT infrastructures.
Some CIOs don’t know the right questions to ask a potential provider. Others are exploring the move to cloud because it seems like the thing to do (or the board suggested it). A few see the cloud as a panacea for all IT woes.
No matter what category a customer is in, partners must put on their trusted-adviser hats and offer an assessment. There are questions to ask to qualify a cloud vendor. There are also questions to ask to qualify a customer for cloud. Compliance issues alone can disqualify a use case. Legacy systems may not port to a virtual machine properly.
Not every situation is the same; not every solution will be the same.
The first step is to dig in and get a holistic understanding of what the business is trying to accomplish and which outcome will be most satisfying. In many cases, we conduct cloud-readiness reviews coupled with IT risk/cybersecurity assessments to craft solutions that make sense for both IT and business unit decision-makers. Those reviews require more depth than is gathered in a standard request for proposal. It may be that the customer has a colocation partner and thinks cloud will save money, but after careful consideration we’ve found that some software won’t port well, control will be lost or costs will rise. There are plenty of other unforeseen factors that will hinder a cloud migration.
As a partner, you need to bring skills to the table. Few customers can afford to have a network architect or Cisco CCIE map out the environment. Make sure you have resources to answer questions including: Which systems should remain on premises or in colocation, and which can be outsourced to AWS or Azure? Are there compliance or security concerns? Who is handling data backup? How is it handled? Is there redundancy? Should we institute data duplication? How critical are certain applications? Is there a service-level agreement, and what is the uptime?
|5 Innovative Ways to Make Money After a Cloud Conversion: Once you have a customer settled in AWS or Azure, it’s time to stack on new services. Register now to find out how!|
You need the right tools and services to address all these questions. For some agents, the answer is to hire a firm to conduct cloud and risk assessments for clients. Many are relatively costly, but nothing is costlier than an extended outage or loss of data. For example, at my company, COLOTRAQ, we have subject-matter experts who can conduct these reviews and assessments, and we extend these resources to our channel partners.
Remember that there is no one-size-fits-all in the world of data-center infrastructure. In fact, I see many CIOs deploying certain elements of their IT infrastructures in physical environments, others on the cloud, and everything in between. The “in between” option seems to be the most widely adopted at this point. Hybrid, multi or private cloud strategies provide many of the advantages of cloud deployments, often without the security and availability risks. In our experience, a mixed strategy is also more cost-effective than a full public cloud deployment.
In most cases, we have found that enterprise customers are not deploying on any platform exclusively across the board. When it comes to mission-critical applications, they still seek to maintain infrastructure in a colocated data center. Other applications and dev environments will be deployed on the cloud, and other types of applications will sit in a managed hosting platform. It’s your job to guide them through to a cloud future.
Dany Bouchedid is CEO of Colotraq, a full-service sourcing and consulting firm that helps companies and institutions source data center infrastructure services such as colocation, managed hosting, cloud computing, network infrastructure and other related services from over 400 service providers worldwide.