Reducing risks with multiple cloud service providers

Using multiple cloud providers can help mitigate public cloud risks, though the challenges of a multi-cloud environment take some getting used to.

IT administrators often worry about losing control of data placed in the cloud. Will data be easily accessible, especially in the event of a cloud provider outage? What if that outage lasts for an extended period of time? A common way to mitigate risk is to employ redundant, independent systems. Public cloud users can achieve this by using two or more different cloud providers.

Using public cloud has both technical and business risks, and both should factor in to a multi-vendor strategy. Technical risks such as outages and other service failures are often at the forefront of IT concerns. Service problems are hardly limited to public cloud providers; issues at corporate data centers are common, though not usually newsworthy. Working with multiple cloud providers or building a hybrid cloud environment helps mitigate the risk of service disruptions in local data centers and with public cloud providers.

Using public cloud has both technical and business risks, and both should factor in to a multi-vendor strategy.

Spreading resources across multiple data centers from the same cloud provider also addresses technical risks. If you deploy instances in multiple data centers of a single cloud provider you can realize comparable redundancy to that from multiple data centers and multiple cloud providers. This, however, does not lower the business risks of using public cloud.

Business risks relate to problems in the professional relationship between your company and a public cloud provider. Problems such as billing disputes or a disagreement over the interpretation of a contract can cause rifts in your relationship with a cloud provider, which may lead to service disruptions. If critical data and applications are dependent on a single cloud provider, you may feel constrained in your ability to negotiate through business disagreements with that provider. In a worst-case scenario, the provider may block your accounts and prevent you from running your software or accessing corporate data. In this case, working with multiple cloud providers gives you more flexibility for both negotiations and data access.

The case against using multiple cloud providers

Despite the benefits of mitigating the risk of data loss, working with multiple cloud service providers can introduce logistical problems. Open source or vendor-agnostic tools and cloud management services can help solve these issues.

Not all cloud providers offer the same services. Storage and virtual machine (VM) instances are often standard, but services such as messaging or workflow and administration tools may vary across vendors. Working with multiple cloud providers may force you into working with the lowest common denominator of services. You can sidestep this problem by choosing vendor-agnostic software applications that will run in all of your clouds. For example, rather than using Amazon's search service, you can deploy the open source search tool Solr.

Technical risks such as outages and other service failures are often at the forefront of IT concerns.

Even when you use services common to multiple cloud providers, there may be differences in how to administer those services. Comparable tools for administrators looking to script cloud administration tasks include Apache Libcloud, which offers a common API for working with cloud servers, cloud storage and load balancers across multiple providers.

Cloud management services such as RightScale and enStratus take service abstraction a step further by offering a consolidated management platform for deploying resources across multiple providers. In addition, they can provide server and application monitoring, access controls, and enforce policies. Cloud management services are especially important in companies using multiple cloud providers that charge resources to multiple budget lines. You might, for example, restrict your data analysts to a monthly budget for computing resources. Rather than devise an arbitrary budget split among cloud service providers, you can set a budget for all computing resources across multiple vendors. Because data analysts would launch all VM instances from the same management console, regardless of which public cloud provider is used, you can track and control spending across providers.

Both advantages and challenges exist when working with multiple cloud service providers. Some of the difficulties, such as the lack of a standard cloud API, are being addressed by open source tools and cloud management services. Avoiding vendor-specific services may seem like a constraint at first but may be a good practice to avoid vendor lock-in. When it comes to the risks associated with cloud computing, using multiple public cloud providers can be a good option.

Dan Sullivan, M.Sc., is an author, systems architect and consultant with over 20 years of IT experience with engagements in advanced analytics, systems architecture, database design, enterprise security and business intelligence. He has worked in a broad range of industries, including financial services, manufacturing, pharmaceuticals, software development, government, retail and education, among others. Dan has written extensively about topics ranging from data warehousing, cloud computing and advanced analytics to security management, collaboration and text mining.

Dig Deeper on Multi-cloud strategies