Every process has its early stages. The same goes for migrating to the cloud. You start with targeted investigation and experimentation to test the waters in an attempt to determine if the cloud is a good fit without requiring significant investment.
When you do start migrating to the cloud, you'll quickly learn you can enjoy such benefits as...
- Agility: the ability to quickly move in and out of resource configurations to address business requirements
- Elasticity: the ability to quickly scale resources out and in to handle spikes and lulls in demand; resources are provisioned just-in-time and fully utilized
- No up-front capital expenditures (CapEx) or long-term commitments; pay as you go
- Increased innovation by lowering the cost of experimentation, and of failure
- Digital transformation through improved employee access to data and collaboration, mobility, and the ability to drive organizational decisions based on analysis of large data sets.
Before you can enjoy those benefits, you have some homework to do first. That's where investigation and experimentation phases of migrating to the cloud come into play.
Cloud Adoption Journey: Investigation and Experimentation
Investigation and experimentation crucial phases of your cloud adoption journey. Maybe your organization is just getting started doing this legwork, or you're already in one of these phases.
When you're in the investigation phase, you typically haven't started migrating to the could and use traditional on-premise client/server infrastructure with local and off-premise redundancy. This could include high availability and disaster recovery clustering. While there may be no “pressing needs" for migrating to the cloud, executives might be feeling competitive pressure to innovate and modernize but might not be sure how to accomplish this. You also might be considering if you want a single-provider option, such as if you should migrate to Google Cloud Platform vs. another provider, for example. Or maybe you are examining the benefits of multi-cloud strategy.
Organizations that have reached the experimentation phase are often eager to learn what migrating to the cloud can do for them. They might be facing a license or service contract renewal, struggling with an inability to scale their infrastructure, working around a lack of features, or have a market opportunity that requires action. Executives might feel pressure to make the “right” decisions regarding a cloud migration.
With no up-front CapEx or commitments — and pay-as-you-go pricing — the cloud provides an excellent opportunity to experiment with different resource configurations so you can “try before you buy.d” If a configuration doesn’t work, you can quickly rip and replace it. This level of experimentation simply isn't possible with traditional infrastructure.
What You Should Be Migrating to the Cloud
It is prudent to move non-core workloads to the cloud to free up time, resources, and capital to focus on the core customer proposition. The following workloads, listed from simple to more complex, allow you to investigate and experiment with the cloud and prove the concept as a first step to adoption.
- Application development, test, and staging environments
- Data storage, archival, backup and recovery
- Cloud-ready web applications
- Disaster recovery and business continuity.
Application Development, Test and Staging Environments
Application development and test environments are no-brainer workloads. The cloud offers improved development and test agility by enabling experimentation with different environment/resource configurations. Often times, test environments will fall under the free service quotas of the cloud provider. These environments offer the ability to eliminate capital expenditures because you can rip them down when you’re done, or spin down the VMs when you’re not using them, paying only for the time you use.
The cloud also offers a unique opportunity to spin up short-term staging areas. These are key when performing application and data migrations, as well as for copying the environment stack for promoting environments from dev, to test, and then from QA to production.
Data Storage, Archival, Backup and Recovery
Cloud storage offers a low maintenance and cost-effective alternative to managing the traditional infrastructure for data archival. Cloud storage providers offer low costs per gigabyte per month for object storage, as well as reduced cost long-term storage with reduced durability. Thanks to economies of scale due to their sheer size and global presence, included are levels of redundancy and availability SLAs that are extremely difficult and costly to achieve within traditional infrastructure environments. Readers are encouraged to consider cloud storage for its reduced cost and maintenance.
Cloud-Ready Web Applications
It is extremely difficult to provision resources for applications with unpredictable, on-off, and bursting demand using the traditional infrastructure. Imagine short-term marketing or other campaigns, seasonal products, and expansion into new markets. It’s easy and costly to over-provision resources based on anticipated demand, as well as time-consuming and potentially harmful to the business to run out of capacity when under-provisioning.
The cloud is ideal for operating web applications. It offers improved agility, scalability and redundancy at a lower total cost of ownership compared with traditional infrastructure, particularly for inconsistent workload patterns. You can spin up VMs on demand within seconds based on images, run the workloads, and then spin them down to ensure resources are provisioned just-in-time.
Platform as a service (PaaS) tools such as Google AppEngine and Managed VMs remove the complexity and maintenance of configuring VMs for scaling, load balancing, and redundancy by managing all these for you, allowing you to focus on your application code. Think web applications such as the corporate intranet, workflow tools, document/help wiki, content delivery, marketing or other campaigns, or e-commerce.
Disaster Recovery and Business Continuity
Disaster Recovery (DR) and Business Continuity (BCP) can be more complex, but they offer significant data loss prevention and loss of use risk aversion. This is done through a “hybrid” environment, one in which some data and infrastructure resides in an on-premise data center while other data and apps in the cloud.
Organizations considering a hybrid environment may be seeking to contain DR costs, have no current DR plan in place, are not happy with existing DR strategy, or have distance requirements for DR environment locations.
There are several scenarios for application and data failover, each with varying levels of complexity to implement. Today we’ll focus on backup and recovery and “pilot light” scenarios.
Cloud backup and disaster recovery involves copying application data to the cloud so that it can be easily recovered and accessed upon failure or disaster. In this case, recovery is manually-intensive — not automated — and incurs a longer recovery-time-objective (RTO). However, it is considerably easier to implement.
Pilot light is unique because it allows you to take advantage of the cloud’s pay-as-you-go model while enjoying a reduced RTO such as faster recovery. In a pilot light scenario, application data is replicated to the cloud, similar to backup and recovery. However, applications are stored on VM images in cloud storage, and may be spun up in minutes to launch the application stack in the cloud. This allows you to pay computer resource usage fees only while the applications are running in the cloud, while paying for storage costs only during hours of normal operation.
Businesses are leveraging cloud technologies to facilitate scaling and growth by migrating non-core workloads into the cloud so they can focus on core business activities instead of managing IT infrastructure. They also can move workloads to the cloud — without needing to make a significant financial investment.
This article is has been updated from the original, which was published on Feb. 19, 2015.