Moving to the Cloud? How to Avoid These 3 Common Cloud Migration Pitfalls

If you’re planning to move workloads to the cloud for the first time, it’s likely you have some concerns based on what you have read or have heard directly from peers. Words that may come to your mind are: time-consuming, complicated, painful, confusing, guesswork, and trial and error. Likewise, if you’ve already migrated some workloads to the cloud and are planning to move more, these words may describe your own experiences.

Many have accepted that this is just the way it is – the cloud offers huge cost and performance benefits, so we’ll just have to suffer through the painful migration process. In actuality, this is NOT the way it has to be. Leveraging the right analytics to inform your decision-making will enable a smooth move to the cloud. These analytics are not just a nice-to-have, but absolutely essential to getting your migration right and avoiding the common cloud migration pitfalls. In the following, I’ve listed these pitfalls along with the analytics that will help you dodge them.


Pitfall #1: Running Before You Walk

A failure in the first stage of a major cloud migration effort has technical, psychological and political implications, which is why understanding which workloads to pilot and migrate in the initial phases is critical to setting up the success of your overall migration. The key is to migrate your least complex applications. The following analytics should be taken into consideration together when determining the complexity of each of your applications:

  • Application Performance: Identify the compute, storage, and network resources consumed by your applications. The machines consuming the least resources should be migrated in the beginning phases because they will have less impact on the rest of your applications and will take less time to migrate.
  • Application Classification: Group your applications based on classification, such as business intelligence, security, or IT Management. Application classes that are not mission-critical should be moved first, while those that are highly critical should be moved later after you’ve already tested the cloud waters.
  • Application Environment: Non-production environments like development, staging, and disaster recovery are best to move early on when you’re just getting started with migration.
  • Application Dependency Mapping: For all of your applications, determine the other applications and servers they are communicating to, how often they communicate, whether the communication is bi-directional or unidirectional, and what the allowable latency is in communications between each. Uncover multi-tier dependencies, as well as dependencies both within and outside of your business unit. Those with the most dependencies, particularly multi-tier and outside your unit, are more complex to migrate.


Pitfall #2: Bad Breaks

Unfortunately, many companies experience application breaks during migration. This results in a massively time-consuming trial and error process. If you don’t have details on each of your applications you will inevitably group them incorrectly and migrate them out of phase. This will increase the number of applications that break during migration and compound de-bugging issues once in the cloud. Consider the following data to avoid bad breaks and get your migration right on the first try:

  • Application Dependency Mapping: In addition to determining application complexity, dependency mapping is also essential for properly building your application move groups. You must move applications communicating with each other together. If you don’t, this communication will break once the applications are migrated, causing serious performance issues once in the cloud.
  • Shadow IT: If your applications are talking to unknown servers, they must be incorporated into your migration plan. Typically, companies have 15% or more Shadow IT and without identifying these, you’ll not only have breakage issues, but also security issues.
  • Firewall Rules: Firewall rules let you know which applications are talking on which ports and which ports you need to open for inbound/outbound application connectivity to work.


Pitfall #3: Budget Mishaps

All too often companies are forced to halt their cloud migration execution because the costs of the initial workloads they moved to the cloud quickly become out of control. There are a few things you need to do to ensure you don’t quickly blow right past your cloud budget. First, accurately measure how much a workload will cost in the cloud before you migrate it to ensure you only migrate those workloads that fit into your current budget. Second, identify the best-fit cloud configuration to migrate your workload to. And third, begin actively monitoring your workloads as soon as you migrate them to the cloud. The following are the analytics you need to achieve all three and avoid cloud migration budget mishaps:

  • On-Premises Workload Performance Profile: For at least two weeks, analyze the peaks and valleys for your on-premises workload’s CPU utilization, RAM usage, storage capacity and occupancy, disk IOPS and bandwidth, and throughput.
  • Cloud Configuration Mapping: Based on a workload’s on-premises performance profile, identify its best-fit (right-sized) cloud configuration to migrate it to, which includes its optimal virtual machine, storage option, and network settings. This will allow you to calculate what your workloads costs will be in the cloud before migration. It will also ensure you move onto the right configuration at initial migration – critical for avoiding surprise cloud bills (see how one company saved 40% on its cloud costs by right-sizing its workloads rather than taking a “lift and shift” approach to migration).
  • Cloud Workload Performance Profile and Usage Patterns: Once you migrate your workload into the cloud, actively monitor its performance, which includes the same metrics as its on-premises performance profile, as well as its usage patterns. Both your environment and cloud options are constantly changing, and only through this continual analysis can you adjust your cloud configuration so that your capacity matches your actual usage. This will ensure you’re avoiding the monthly sticker shock and meeting your performance requirements so that you can stay on track for completing your additional workload migrations.


By taking a deep data-driven approach to your cloud migration, you’ll reduce migration time and create a clear plan to guide your migration process, including which applications you will decide to migrate together and the best order in which to move and pilot them.  Only with the right analytics can you design a phased approach to the cloud that aligns to your business needs and escapes those common migration pitfalls. Many organizations have unfortunately experienced these issues because they have taken a manual approach to this data collection and analysis, which is nearly impossible to do right. The good news is that there are solutions available to automate this analysis on an ongoing basis, dramatically improving the accuracy and efficiency of cloud migration.

Learn more about the Cloudamize platform's automated  data analysis for cloud migration & cost management

Related Posts