It feels as if every business these days is competing in the Amazing Race to the cloud. Whether an organization hopes to take advantage of the cost savings that cloud computing can provide or increase overall agility, this “cloud first” mentality frequently leaves today’s IT professionals tasked with managing applications in a challenging environment: the hybrid data center.
Although industry experts anticipate a 44 percent annual growth in cloud-based workloads over the next several years, the reality is that despite rapid cloud adoption, many businesses will continue to have a significant part of their daily operations tied to physical infrastructure and its processes for many years to come. The Cloud is for everyone, but not for everything. Rather than expend resources to rearchitect applications for cloud deployment, IT teams will most likely “lift and shift” applications originally designed for traditional on-premises infrastructure to the cloud, creating a whole set of unique challenges and potentially confusing management scenario.
In addition, although it’s true that cloud computing offers significant cost savings, increased agility and flexibility and is an especially good test environment for applications, some of the cloud’s limitations can create notable friction points when an on-premises IT team looks to shift into the cloud.
For example, most cloud environments are designed to use commodity, low-cost hardware and are optimized for agility rather than uptime. With most cloud-service providers, the responsibility for uptime is in the customer’s hands, as they embrace a philosophy of “architect for failure,” recommending redundant, multi-region, self-healing deployments.
Without the proper research, an unsuspecting administrator might move a mission-critical workload to a cloud environment that is not designed to provide the level of uptime or security required, leading to a myriad of performance and data compliance issues.
Alternatively, the ability for any administrator to easily provision a new cloud environment using either a corporate or personal credit card is a double-edged sword. Yes, test environments for in-house developers can be generated at the drop of a hat, but that ability also presents serious governance and security issues. If the administrator managing that environment suddenly leaves the company, he’s taking the keys to that data with him.
Ultimately, unless it’s a brand-new application architected from the get-go to be cloud aware, administrators are essentially tasked with running an app in the cloud that was designed to run in a non-cloud environment and that has many of the same properties as one built for on-premises operation.
More importantly, because hybrid cloud is most likely a multiyear journey, more applications and more components of applications are likely to be moved to the cloud over time. So, the question of the day becomes, “How do I manage these applications using the same tools and methodologies as I am managing on-premises apps?”
To that end, there are several best practices that can be implemented to align the management of on-premises and cloud-based applications—and yes, these do involve a good dose of “cloudy thinking.”
Application and End-User Focus
Even in a hybrid environment and in relation to applications still running on-premises, there is an on-going trend to focus less on infrastructure components and department silos, and more on end users and the applications. Infrastructure serves only to make the application work. Cloud infrastructure is ephemeral and dynamic.
To achieve some of the resiliency and flexibility that the cloud offers, administrators should note that in the cloud, servers are disposable—if one begins to fail or underperform, it’s time to kill it and bring in a new one. IT should also consider ways to implement shorter, more-iterative processes to allow teams to move faster, innovate and serve the business more effectively. And it’s important to not forget that automation of code, deployments, tests, monitoring and alerts goes a long way to faster end-user servicing as well.
Collaboration
Whether an application is supported on-premises or being managed in the cloud, the ultimate objective is to provide the end user with peak application performance, and as such, silos do not work. Today’s IT mindset should increasingly be that if an application is down, everyone is responsible. There is no database, virtualization or storage team. There is only the IT team, and they are responsible for the performance of applications. This approach requires transparency, visibility, a consistent set of tools and teamwork.
The opportunity is to align the entire IT team behind the same the goals and metrics: maintaining application uptime and improving the end-user experience. Having a common goal increases collaboration and is likely to increase the business value produced.
Research
Although not necessarily a “cloudy thinking” principle, research should be a consideration for any IT team in the process of evaluating a cloud provider. Administrators tasked with managing hybrid environments must do sufficient due diligence with regard to a cloud provider’s processes, SLAs, recommended architectures and available capabilities—does it provide the level of security they need? Is the stated percentage of uptime realistic and reliable? How resilient is the environment? This should also include the implementation of governance rules for IT and the organization to prevent data compliance issues.
These forward-thinking SLA questions and governance processes will help make the move to a hybrid data center more seamless and ultimately affect how successful an admin will be when managing applications in the cloud.
Full-Stack Monitoring
Moving workloads to the cloud—even a few—does not happen in one day. In some cases, IT may move just parts of an application to the cloud. Regardless, a successful transition requires a reliable, full-stack monitoring system. To ensure optimal performance, everything should be tested, and the impact of every change known. The cloud allows more flexibility, more control and instant changes—all of which are reflected in your monthly bill. It creates a responsibility and an opportunity to optimize everything. When managing a hybrid environment, using a comprehensive tool provides visibility across the stack into the health of not only physical infrastructure and cloud environment, but also the performance of any application is critical for speed, collaboration and end-user satisfaction.
Although these best practices were born in the cloud, they are also useful when applied to traditional on-premises IT. Even in a “lift and shift” scenario, the goal of moving certain workloads to the cloud is that IT departments can begin using cloud-based principles and some of the new available services more frequently, providing them with benefits that today’s “cloud native” players are enjoying. With these best practices in mind, administrators can begin to blur the line between management of on-premises and cloud-based applications, and ultimately allow for a faster, more agile business.
Leading article image courtesy of BasicGov under a Creative Commons license
About the Author
Gerardo Dada is Vice President of Product Marketing and Strategy for SolarWinds’ database and applications business globally, including SolarWinds Database Performance Analyzer. Gerardo is a technologist who has been at the center of the web, mobile, social and cloud revolutions at companies such as Rackspace, Microsoft, Motorola, Vignette and Bazaarvoice. He has been involved with database technologies from dBase and BTrieve to SQL Server, NoSQL and DBaaS in the cloud.
The post Hybrid Infrastructure: An Application Collision Course? appeared first on The Data Center Journal.