Prepare for Success in Cloud Migration: Elevate Above Infrastructure and Silo Tools
November 27, 2018

Eric Kraieski
Transitional Data Services

Share this

Achieving success with cloud adoption remains an elusive challenge for many organizations. But why is that the case? After all, there are countless tools designed to facilitate the process of taking on-premises operations to the cloud. It is common to use these purpose-built tools for moving virtual images, automatically provisioning services, migrating data, right sizing deployments and optimizing cloud operations. But when it comes to application migration, this variety of infrastructure tools actually contributes to the problem.

These disparate systems certainly work well enough for their specific use and purpose. However, successful execution of application rehosting requires that users look above these infrastructure tools to see the full picture and select the tools appropriate to the specific migration method, or "R approach" — rehost, replatform, repurchase, refactor, retire, retain — for the application.

Perhaps the most interesting thing about these point solutions is that they all focus on infrastructure migration tasks, while selecting the R approach and the specific sequence of steps required are entirely dependent on the application and business goals. How is it possible to decide which R method is optimal or appropriate if you never look into the application requirements? Before committing to a specific path for each application, it is essential to consider the business needs for its availability, access, performance and the total cost to achieve the migration.

Of course there will be times where a straight rehost (often called a "lift and shift") will be appropriate and optimal. This is especially true for mobile workloads that are self contained on a specific server with few external dependencies, no external storage and no requirement for real time app-to-app communications. In other cases, a rehost may be possible, but not desired. For example, rehosting may violate corporate information security policies such as HIPAA or GDPR. Or perhaps some refactoring or reconfiguration is required to achieve target availability goals. Organizations should assume no more than 10-20% of their total server inventory will be highly mobile and ready for an automated rehosting process.

In our experience, organizations that proceed with an "infrastructure first" approach quickly recognize that it does not take into account the impact on critical business applications early enough in the process. Unfortunately, this approach often results in wasted time as plans must shift once the business factors are acknowledged and considered. Taking an application-centric approach is the only way to orchestrate a successful cloud migration.

Know Your Environment Before You Commit to a Migration Approach

Cloud-native platforms typically provide much more agility and flexibility than lifted premise ones. Refactoring apps can be costly and time consuming, so some organizations prefer the lift-and-shift approach. But before you make the decision to take a lift-and-shift migration approach for an application, carefully consider your current environment, your business and IT goals, and expected cost and staff impact first. Simply shifting a legacy, premise-based app to the cloud with all its limitations instead of refactoring it may limit you from taking full advantage of all the benefits associated with the cloud.

Bottom line, tools to support general planning and to assess server mobility can provide useful data points, but are not sufficient for comprehensive planning. Developing appropriate migration plans requires a comprehensive understanding of the entire environment including appropriate information from siloed tools and application SMEs (subject matter experts). Applications should be identified and assessed for readiness: some will be able to be migrated immediately, while others will have to be rewritten or modernized to be workable in the cloud. The indispensable first step for any migration is having an actionable picture of the entire data center, which includes:

■ Accurate information about where apps reside, who owns them, and what SLAs, RTOs, RPOs apply.

■ Complete knowledge regarding the application dependency landscape. Don't just rely on autodiscovery. Your subject matter experts know the ins and outs of your business and will be able to tell you things that machines can't detect.

■ A normalized view of the landscape.

■ Visual dependency mapping of the entire landscape, including what applications are dependent upon, and what is dependent upon them.

■ An understanding of what applications should generally be "grouped together" for a cloud move.

■ The ability to distinguish superfluous data from information that matters. For example, the operating system, manufacturer/model, and IP address are commonly used data points in migration analysis and planning activities, while other information such as CPU speed, MAC address, BIOS, and OS Install Date are simply not necessary or beneficial to the migration activities. Tracking unnecessary data will distract the team and slow down discovery. Don't boil the ocean; just capture the data you need.

Mastering Orchestration

Application migrations are among the most complex projects an organization can undertake and require a cautious approach. If you take the simplest path, assuming that rehost is the preferred approach, then rapid early progress can be achieved by first focusing on the easiest mobile workloads. But once you complete the migration of the easiest workloads, the progress will come to a screeching halt. The majority of your app-to-cloud migrations will require deeper analysis, more careful planning and choreographed execution to assure success.

Orchestration of such an ambitious and sometimes treacherous initiative may seem to be an elusive goal. To avoid mishaps or stalled projects, here are several tips for orchestrating successful outcomes of your cloud migration initiatives:

1. Move up the stack, take an application-centric approach

2. Establish visibility across all silos and users

3. Don't boil the ocean – leverage a sprint-based, iterative approach

4. Leverage existing info and tools where available

With a disciplined approach, you can drive successful outcomes for your cloud adoption initiatives. You'll achieve greater agility and scalability in hosting solutions while avoiding any unplanned outages of your business applications and services.

Eric Kraieski is VP at Transitional Data Services
Share this

The Latest

June 13, 2019

Establishing a digital business is top-of-mind, even more so than last year, as 91% of organizations have adopted or have plans to adopt a digital-first strategy, according to IDG Communications Digital Business Research ...

June 12, 2019

If digital transformation is to succeed at the pace enterprises demand, IT teams, the CIOs who lead them, and the boardroom must forge a far greater alignment than presently exists. That is the over-arching sentiment expressed by IT professionals in a recent survey on the state of IT infrastructure and roadblocks to digital success ...

June 11, 2019

Given the incredible amount of traffic traversing corporate WANs, it's not surprising that businesses are seeing performance issues. If anything, it's amazing applications work as well as they do ...

June 10, 2019

Are your business applications sluggish? Choppy? Prone to getting hung up or crashing at the most inopportune times? If these symptoms sound familiar, you might be suffering from the heartache of … poor application performance. Stop me if any of this sounds familiar ...

June 06, 2019
AIOps Exchange, a not-for-profit private forum defining the future of AIOps, published <span style="font-style: italic;">The AIOps Manifesto</span> discussing the role of AI in supporting digital transformation ...
June 05, 2019

As network transformation initiatives like SD-WAN, edge computing and public/private clouds are adopted at increasing rates, hybrid networks are quickly becoming the new normal for IT and NetOps professionals.Without visibility into these hybrid network environments, NetOps are unable to troubleshoot the business-critical applications every organization relies on today. Here are four ways IT and NetOps teams can gain better visibility into complex, hybrid networks ...

June 04, 2019

A minimum Internet Performance Bar exists that, if met, should deliver top-tier website performance, regardless of industry, according to the 2019 Digital Experience Performance Benchmark Report, from ThousandEyes, a comparative analysis of web, infrastructure and network performance metrics from the top 20 US digital retail, travel and media websites ...

June 03, 2019

Since digital transformation is happening at such a rapid pace based on new, highly complex technologies like multi-cloud, containers and microservice architectures, customers are experiencing more challenges than ever in managing this complexity. However, with every challenge comes an opportunity. So, how can channel partners leverage these market disruptions to open the door to opportunity? The answer is simple ...

May 30, 2019

Executives from proactive organizations reported using performance management strategies to deliver innovation and meet broader business goals, and implementing application performance management (APM) tools with advanced monitoring features such as real-time user experience monitoring, and providing a composite view of log and performance data, according to Driving Business Performance Through Application Performance Management, a new report from GigaOm ...

May 29, 2019

Through our recent study, we wanted to better understand how service desk users are interacting with the service teams; how they connect for service; the manner in which most service desks receive user requests; and if organizations employ a knowledge base and how that information might be stored. Here’s what we’ve discovered ...