Cloud technologies have truly been transformational for virtually every aspect of IT and business. Cloud has enabled new paradigms for the delivery of software and services by offering flexibility, scalability and speed of deployment for both large and small businesses. However, there continue to be several misconceptions about cloud computing that are still pervasive and need to be addressed.
Cloud Misconception #1: I am less secure running in the cloud than in my own datacenter.
Security concerns are still the one of the most pervasive misconceptions that shows up in survey after survey of why small businesses are hesitant to move to the cloud. The misconception that business will lose control of their data and be exposed to higher risks if running on cloud infrastructure is still commonly held. Sure, you still need to be diligent about security when running in the cloud to protect against hackers, viruses and poorly educated end users performing actions they should not, but cloud can actually help you increase your security in many instances.
The reality is that most cloud vendors have far greater security measures in place than most businesses can achieve on their own. For physical security, cloud facilities are typically setup with 24x7 physical security including round the clock guards, biometric access control through the facility and electronic surveillance. Likewise, cloud vendors are able to hire IT security specialist to deploy and monitor the latest in multilayer security systems to attempt to thwart hackers and bots much more cost effectively than a small business IT team.
In addition, cloud vendors undergo regular security and procedural audits to make sure they are meeting the latest standards for security. Ask for your cloud vendors SOC 2 report if you have questions around the level of security they are maintaining.
Cloud Misconception #2: I don’t need to worry about backup with cloud. The cloud vendor takes care of that for me.
The fact is that cloud vendors, including SaaS applications like Office 365 or Google G-suite, protect their infrastructure from downtime. They don’t protect your data at the level you probably require. You are responsible for protecting your data in the cloud.
Microsoft and other SaaS vendors are very clear in their Service Level Agreements that customers are responsible for their own user data. If user data is lost due to accidental deletion, malicious actions, or ransomware then you are responsible for restoring your lost data. The Office 365 Service agreement states, "We recommend that you regularly backup Your Content and Data that you store on the Services or store using Third-Party Apps and Services.” If you cloud data is important to you make sure you are protecting at the same level as you were when it was in your datacenter with backup tools.
Cloud Misconception #3: Cloud infrastructure is more reliable than standard servers.
This misconception is based on a common misunderstanding of how cloud infrastructure is typically built. There is a frequent statement that is made that “Cloud is really just using ‘someone else’s’ computer.” However, cloud infrastructures are built very differently than the servers that most businesses run their applications on today in their server closets. This results in fantastic benefits such as greatly increased flexibility and agility but creates new challenges for many application that are not cloud-aware.
Cloud architectures are built at a scale and a level of automation not typically seen in most enterprises and never seen at SMBs. As such, hyper-scale cloud infrastructure in particular are designed to fail. In a traditional server, when a hard drive begins signaling failure or a redundant power supply dies, IT admins will rush to repair that component to get the system back to 100% health. In cloud, the failed component is simply tagged as unavailable and its workloads or data are simply migrated to a new node. Cloud native applications are designed to take these failures and migrations into account. Legacy application running in the cloud typically do not and can therefore experience downtime as a result.
While it is certainly possible to build robust, highly reliable cloud applications, it usually requires refactoring or rearchitecting the application to consider that cloud component can fail without notice. Indeed, native cloud applications are often built from the ground up with failure awareness and are designed to automatically restart workloads on new compute nodes to keep running and run stateless as much as possible. If you are planning on simply moving legacy applications as is, then cloud may cause you more downtime than you had previously.
Cloud Misconception #4: If I account for the Compute (number of CPU cores/amount of memory) and Storage (# of GB of of storage) I am using, I will know my cloud budget.
Cloud budgets often surprise new cloud users. Cloud infrastructure and applications typically bill using a utility model, where you pay for what you use. This result in more efficient spending by eliminating the need to plan for peak usage and pay for unused capacity. In fact, this increased agility to increase or decrease cloud usage is one of the biggest benefits of moving to cloud. However, many new cloud users get surprised by fees they were not expecting. Compute and storage expenses are typically easy to estimate up front. However, many new cloud users don’t expect to pay for networking fees on a per GB basis or are surprised at high quickly there networking fees can add up. For example, hyper-scale clouds like AWS and Azure charge network egress fees to any data that leaves the cloud or transfers between data zones. At up to 15 cents per GB, those egress fees can really add up.
Want more guidance on how to articulate where cloud is heading? Download our Explaining the Cloud to Your Clients eBook today.
By Lily Teplow
By Brian Downey
By Dave LeClair