Don’t Even Think About Cloud Migration Unless You’ve Done This First
Posted by on May 18, 2017
Chuck Darst is a Senior Product Marketing Manager for Cherwell Software. Chuck has over ten years of IT Service Management (ITSM) industry experience and over 20 years in a variety of IT Operations Management (ITOM) roles with a focus on machine learning, automation, compliance, and IT security.
It’s hard to find a tech article that doesn’t reference the continued adoption of cloud computing services. Building a new application or introducing a new service offers a rare opportunity to take advantage of new methodologies—containers, microservices, continuous deployment, and so on—the cloud has to offer. But, what about moving an existing large, composite application or service to the cloud?
A major software upgrade, physical data center move, or equipment refresh/upgrade all provide opportunities to evaluate the role the cloud plays with respect to your overall infrastructure. While there are a number of considerations that may influence your cloud migration strategy, one thing is clear: it is important to understand all the components that comprise any given application or service prior to migrating. This includes network connections, storage, application servers, web servers, database servers, related software processes, security elements, and possibly more.
Application dependency mapping provides visibility into these relationships to help ensure that nothing is overlooked.
Identifying Move Groups
A traditional use case for application dependency mapping (ADM) has been identification of “move groups” which have been used in data center consolidation and migration projects for years. Move groups are also commonly used in disaster recovery planning and remediation programs.
When moving an application or a physical section of a data center, it is essential to understand all parts that make up the application(s) along with the dependency relationships, or “what is connected to what.” This is especially true for distributed or composite applications. In an infrastructure move, you don’t want to break an application that has a process running on a physical or virtual machine located outside of the move group. This same information is used in disaster recovery planning to identify all physical and logical components that need to be restored.
ADM provides insight into the software processes and applications running in a physical section of a data center. Likewise, ADM provides visibility into all physical and logical components that make up an individual application or service. A move group can then be identified for either a physical or logical application move—ensuring that everything that needs to move is included.
The same concept can—and should—be applied to cloud migrations. Moving an existing application or service applies in particular to cloud-based infrastructure-as-a-service (IaaS), but could also include platform-as-a-service (PaaS) or even potentially software-as-a-service (SaaS) elements.
Targeting Resources for Decommissioning
ADM also helps identify servers and processes that don't serve a purpose and should be decommissioned—ghost servers that are running, consuming power, taking up space, leased, or possibly have applications and software licenses running that nobody uses. If you’re planning any type of application migration or data center consolidation, now is a great time t decommission or re-purpose these systems.
This applies to both physical data centers and cloud services—especially given the ease of provisioning a cloud service intended for a short term use and the potential to forget about it.
Optimizing Cloud Management
Once a cloud migration has occurred, ADM provides an important ongoing benefit of keeping configuration information accurate—improving Event, Incident, Problem and Change Management processes, especially in more complex and service-centric environments. ADM also helps detect configuration drift which has audit, compliance, and/or even security ramifications.
To be clear, ADM generally won’t have the same infrastructure depth below the virtual machine or equivalent level in cloud-based environments as it will within the physical data center. What constitutes discovery and how information is kept current in a cloud can also be different, especially in more dynamic service environments. Regardless, ADM continues to provide insight into relationships that results in more effective infrastructure management and cloud service strategy.
Next Up: Cherwell Application Dependency Mapping
Minimize T infrastructure risk and improve change outcomes