(c)iStock.com/Barcin
By Ryan Kroonenburg, Managing Director at Logicworks UK
After speaking with hundreds of UK technology leaders about cloud adoption, it is clear that cloud technology is transforming business models and improving cost-efficiency in the enterprise.
Despite the positive results, IT leaders have also shared serious concerns. They usually sound something like this:
“The cloud is great. The support is mediocre at best.”
“I am not sure my cloud provider understands my business.”
“I spend thousands of pounds a month for cloud support, but I do not know what they do.”
Enterprises expect a certain level of support for IT products, and they are simply not getting it. In fact, nearly seventy-five percent (75%) of UK CIOs feel they have sacrificed support by moving to the cloud. Eighty-four percent (84%) feel that cloud providers could do more to reduce burden on internal IT staff, and the vast majority of respondents felt ripped off by “basic” cloud support.
This is a huge threat to the success of the cloud in the UK. Poorly architected clouds that are supported by junior, outsourced technicians not only expose enterprises to downtime and security vulnerabilities, but also hurt overall business goals and reduce the likelihood of further cloud adoption.
I believe there are three major causes of cloud support failure — and in many ways, it is up to enterprises to become educated and demand a higher quality of service.
Cloud washing
While this phenomenon is well-known in North America, the UK market is still full of small service providers that claim to provide “cloud computing” but in fact do not. Usually these ‘cloud providers’ are small companies with a couple of leased data centres that provide none of the scale, built-in services and tools, and global diversity that true cloud providers offer.
These companies have a tiny fraction of the power and capacity for innovation of cloud giants like Amazon and Google. Cloud technology has far outstripped the basic low-cost compute model, and has now transformed data warehousing, analytics, cold storage, scalable compute, and more. Small cloud providers cannot possibly compete.
UK IT leaders assume that these small providers are superior in cloud support and cost. Unfortunately, the typical support model tends to be “fix what is broken”, which is insufficient for cloud systems and results in slow wait times and higher risk of manual error. Further, these niche providers lack the scale and breadth of services to keep costs down.
Misaligned support models
When your applications depend on the health of physical data centre components, you want a support team that is very good at fixing mechanical issues, fast.
When your systems are in the cloud, you want a support team that builds systems that can survive any underlying mechanical issue.
These two infrastructure models require very different support teams. In the cloud support model, the system as a whole is more complex. For instance, a problem might originate in a script in a central repository, so the support engineer must have a deep understanding of all layers of the system in order to discover the source of the issue. They must be able to code as well as to understand traditional networking and database concepts. Service providers can no longer staff support teams with low-level engineers whose only responsibilities are to record issues and read monitoring dashboards.
Unfortunately, cloud service providers — or traditional hosting providers that have rebranded — do not staff their support centers with experienced cloud architects. It is no surprise to me that nearly half of IT leaders report that call handlers lacked sufficient technical knowledge (41%) and were slow to respond (47%).
The cloud automation gap
There is only one way for service teams to deliver a fast, targeted fix to a service request: cloud automation. Not incidentally, automation is also the only way to deliver 100% available systems on the cloud.
Maintaining complex cloud environments is difficult. To take full advantage of the cloud’s flexibility and pay-as-you-go pricing, your cloud should scale dynamically without human intervention. This requires more than just a basic service install: you need to automate the provisioning of new instances, which means configuring those instances quickly with a configuration management script like Puppet. If your team’s goal is to deploy more frequently, you need to combine this rapid infrastructure provisioning capacity with deployment automation tools, which automate the testing and deployment of new code.
When small cloud companies claim to offer scalability, be sure to dig deeper. What they may mean is that they can manually respond to a service request to increase your server capacity. If you want to actually automate scaling on a large cloud platform, these tasks require advanced, specialized skills both to create these systems in the first place and to maintain them.
Few service providers offer true automation, and few enterprises realise they need it. Automation is difficult, automation experts are hard to find, and this education gap will have even larger consequences in the years to come. As the demands on IT resources continue to increase, the delay required to perform or outsource manual work will become impossible to sustain. Downtime is impossible to prevent when the system cannot be adequately automated for failover.
Any cloud service provider must be cloud automation experts. Cloud automation should be the heart of enterprise support — it will decrease downtime, increase flexibility, and dramatically improve your provider’s ability to rapidly respond to service requests.
The right providers
Choosing the right cloud platform and service provider feels very risky for many UK business leaders. The industry is changing rapidly. The pressures to develop faster and more cost-efficiently are enormous.
I place my bets behind companies that are also evolving rapidly and developing the right new services to meet customer demand, which is why I have helped European enterprises move to AWS for over five years. That’s also why I joined Logicworks, who I believe are one of the only cloud service providers to understand the importance of automation in improving data privacy and agility. Feel free to reach out to me directly (rkroonenburg@logicworks.net) to learn more about AWS or Logicworks.
