Public IaaS and PaaS clouds have existed long enough that they have become mainstream technologies. Many organizations have adopted a “cloud first” IT model, in which the cloud is the default location for deploying new applications and services.
Given this trend, the focus on the cloud has largely shifted from cloud adoption to implementing multi-cloud deployments. This article explains why multi-cloud infrastructures are useful and how to implement them.
Although multi-cloud environments are undeniably more complex than single cloud environments, there are compelling reasons for adopting a multi-provider approach to the cloud. One such reason is that using multiple clouds allows organizations the flexibility to choose best-of-breed solutions. There are certain services that nearly all of the major cloud providers offer. For example all of the major public cloud providers offer an object storage solution, a platform for hosting virtual machines, and a database service.
Whatever the service, there is a good chance that one provider will offer a better solution than the others. One provider might, for instance, have a better virtual machine platform, while a different provider may have a better storage solution. Using multiple providers gives you freedom of choice.
Just as operating in a multi-cloud environment gives you the ability to choose a best-of-breed solution from a number of competing services, multi-cloud environments also give you the ability shop for the best price. You might find that one provider has the best price for database services, while a different provider has the best price for virtual machine hosting.
Security and redundancy also tend to be strong considerations when implementing a multi-cloud environment. As the old saying goes, you shouldn't put all of your eggs in the same basket. An organization might, therefore, mirror a database to a different cloud provider, so that a copy of the data will remain available in the event that the primary cloud provider suffers outage.
Likewise, multi-cloud usage can help to keep data secure (or private). Some organizations use erasure coding to stripe data across multiple cloud storage platforms. Very often, the data is written in such a way that no one single provider has a complete copy of the data. Instead each cloud contains only fragments of data that are useless by themselves. Hence, if a cloud provider is compromised, it will be impossible for the hacker to read the organization's data.
Supporting a Multi-Cloud Environment
There are two main things that your organization will need in order to support a multi-cloud environment. First, you will need the appropriate staff resources. Each cloud service provider has its own way of doing things, and so expertise in one cloud environment does not necessarily translate to expertise with another.
Simply put, the members of your IT staff will need to have a good working knowledge of the cloud platforms that are being used. Not every IT staff member needs an intricate knowledge of every cloud platform, but the IT department should collectively have the knowledge required to support each cloud that is being used.
The second requirement is having a strategy for moving resources to a different cloud. There are actually two different considerations involved in this requirement. The first consideration involves the actual migration of data. The second consideration is being able to use the data once it is in its new location.
Obviously, there is no universal answer to these requirements, because each organization's needs are different. When it comes to data migration, however, the best option is to check to see what types of data migration tools the public cloud provider makes available to its subscribers.
Alibaba Cloud's Support for Multi-Cloud Architectures
Alibaba Cloud provides a number of different tools for uploading and downloading data, which can be useful when implementing a multi-cloud architecture.
While you may not need every one of these tools, different tools are useful in different situations. Alibaba Cloud's tools are listed at https://www.alibabacloud.com/help/doc-detail/51654.htm and are divided into two categories-Alibaba Cloud DTPlus products and open source products. The Alibaba Cloud DTPlus products include tools such as Data Integration of DataWorks (a data synchronization tool), MaxCompute Client (a data uploading and downloading tunnel), and Data Transmission Service (DTS). DTS is a data migration tool that is designed to work with RDBMS, NoSQL, OLAP, and other data sources. The open source tools are available on GitHub and include Sqoop, Kettle, Flume, Fluentd, LogStash, and OGG.
There are a number of compelling reasons why it may be beneficial for an organization to adopt a multi-cloud strategy. When doing so, however it is important to carefully assess the various public cloud providers in order to determine which providers will best meet the organization's own unique needs. One of the best ways of assessing a cloud provider's capabilities is to sign up for a free trial. Currently, Alibaba Cloud is offering a free trial: https://www.alibabacloud.com/campaign/free-trial#free-products. This free trial includes $300 of credit, which you can use to explore Alibaba Cloud's various cloud offerings.
Brien Posey is a Fixate IO contributor, and a 16-time Microsoft MVP with over two decades of IT experience. Prior to going freelance Brien was CIO for a national chain of hospitals and healthcare facilities. He also served as lead network engineer for the United States Department of Defense at Fort Knox. Brien has also worked as a network administrator for some of the largest insurance companies in America. In addition to his continued work in IT, Brien has spent the last three years training as a Commercial Scientist-Astronaut Candidate for a mission to study polar mesospheric clouds from space. You can follow Posey's spaceflight training at www.brienposey.com/space