Enterprise Grade Challenges
So how should an enterprise migrate its apps to the cloud? First, keep in mind that you should not start your plan with tightly-coupled legacy apps. Instead, the preferred approach is to stay on the safe side by starting with standalone applications, such as specific mini sites to serve a one-time marketing campaign.
Next is to decide how to migrate compute storage and networking to IaaS, using AWS or Azure for example. This includes mitigating risk by validating critical pieces of your proposed architecture, and identifying existing IT tools that you can use to operate your cloud – such as resource provisioning and configuration, and integration and monitoring tools.
When migrating apps to the cloud, you should consider factors such as network configurations and interdependencies, SLA requirements, and how availability requirements such as disaster recovery and scaling fit into your migration.
Other challenges to consider include understanding your industry and its possible limitations. For example, healthcare organizations that migrate to the cloud may have to take into account strict compliance standards and regulations from the Health Insurance Portability and Accountability Act (HIPAA). This act requires – among other things – that organizations facilitate processes that create and maintain retrievable copies of protected health information at all times.
Steps and Tips
Keep in mind that when migrating, you should first qualify and research your migration strategy. You don’t need to wait to move your production and only then think about protecting your data and apps. You will want to implement this as you run your initial non-mission critical workloads, such as testing/development – so that you don’t lose your development/testing work as well.
The first step in your migration is to keep a list of which apps can be moved and which cannot. To do so, you should list your application catalog and know who is using the apps and how often are they being used. You’ll also need to check workload sensitivity and criticality.
At this stage, you can also select the migration method. Gartner’s past analysis says that when migrating to IaaS, you can either rehost (i.e. don’t make any changes and simply redeploy the apps on an IaaS), or revise (i.e., the current code can be extended or modified to fit to cloud environment).
The next step is to test the workload. Here you should run your proof of concept, as well as tests for validating the performance of the workload and the costs for running on the cloud.
This phase will help you get a feel for your resources and capacity that your applications require – such as storage types and size, the number of VMs needed, and network bandwidth. You’ll also have a better understanding for which storage services you can use to replace your traditional data repositories (for example, using Amazon RedShift to run your data warehousing requirements).
Make sure you test the system in a way that is representative of the final production environment. This will ensure that you know which network and security controls are needed to replace your legacy firewall systems.
Now comes the actual migration phase. You should start this phase by making a blueprint design of your cloud environment. List the migration plan that you will use for automating the move and the list of tools to manage aspects, such as performance and cost monitoring and replication to implement backup and recover when running within the cloud.
The migration plan should also list the tools that will be used for automating the move. Learn which services your cloud vendor has to offer for supporting servers and data migration in a reasonable time (taking into account the amount of data you need to transfer).
The last step is to take your ongoing operations into consideration. You should be able to accept and embrace the frequent changes that migration to the cloud brings. You’ll need to build a transparency and maintenance layer, and then make sure that the layer supports IT management aspects such as performance, cost, security and availability (backup and DR).
Now is also the time to select an appropriate DR scheme, based on failover or fast backup/restore. The ability to roll back a running system to a particular point in time is crucial, so ensure that you have a good cloud backup service that can do so.
Make sure as well that you know and understand what tools are available to you by your cloud vendor. Tools such as the AWS Trusted Advisor and AWS Cost Explorer can come in handy for getting the insight you need into security configurations, costs and usage.
The trend of taking the next data center step in the public cloud is gaining steam, with companies going all-in with AWS and other cloud platforms. You should be aware of the Amazon Shared Responsibility Model, which ensures coverage on the physical security, and means that you are responsible for your migrated applications performing and being secured using AWS infrastructure and various building blocks.
For all workloads – whether on-premise or public – migrations should be treated as a strategic move for enterprise IT. Cloud Protection Manager (CPM) was designed to address all of the backup needs of companies migrating to the AWS cloud today. CPM has become an attractive alternative to traditional onsite backup systems, because it provides flexible backup policies and scheduling, rapid recovery of instances, and a simple, intuitive and user-friendly web interface to easily manage your backup operations. CPM allows users to manage multiple AWS accounts and configure policies and schedules to take automated snapshot backups. With CPM, you can recover a volume from a snapshot, increase its size and switch it with an existing attached volume in a single step.