The Infosys Labs research blog tracks trends in technology with a focus on applied research in Information and Communication Technology (ICT)

« Cloud Migration in Banking Industry | Main | NextGen User Interfaces »

Multi-cloud is the way for cloud adoption

The model of Multi-cloud is mix and matching the best-of-class application, solutions and services from more than one cloud infrastructure providers for creating the most suitable IT landscape for a business.  Deployment can make use of public, private clouds, or some combination of the two. Another aim of multi-cloud deployments is to offer redundancy in case of infrastructure failures and avoid vendor lock-in. In hybrid cloud model, organizations use a combination of public cloud, private cloud and on-premises services, wherein in multi-cloud model employs multiple cloud services from more than one provider. These approaches can coexist, example of an organization that uses a private cloud solution, an on-premises server and different public cloud solutions would have an IT strategy that is both multi-cloud and hybrid cloud.

There are few major drivers from technology as well as from business perspective for the advent of multi-cloud adoption. Avoiding vendor lock-in is key concern for many businesses, and the approach makes it easier to transfer between cloud service providers for pricing or differing capability choices. From performance optimization standpoint properly managed cloud architecture provides the organization freedom to seamlessly shift IT operations as needed. level of resiliency available with multi provider can minimize the risk and effectiveness of DDoS attacks. Improved reliability comes with failover solution of otherwise passive cloud as primary cloud falters processing a service request such as an e-commerce transaction. Also, multi-cloud model allows organizations to effectively track and analyze infrastructure expenditure and utilization and forecast future costs

But there are multiple challenges faced by the organization on the road towards adoption:

·         Getting enough visibility into application landscape with multitude of clusters running may become a challenge; leading to improper consolidation and prioritization of resources

·         Organizations want consistent security policies across all environments, which requires management of configurations and workloads appropriately based on compliance.

·         Managing workloads deployed across multiple disparate platforms creates need for real-time processing and orchestration of services and applications

·         Organizations may lack capabilities such as backing up applications, options for disaster recovery or workloads movement across environments

·         Organizations needs to adopt containerization to build applications that can easily move across environments without having integration difficulties

There are broadly five multi-cloud models conceived by various organizations. One, development in cloud A deployment in cloud B; most common and simplest approach as the name suggests and key factor being cost optimization by rationalization. Two, burst to cloud; services typically runs in on-premises environment, but during peak demand spins up additional capacity into the cloud with the goal of avoiding extra cost hardware procurement. Three, multi-cloud disaster recovery; in this model production application run in one cloud environment with maintaining separate copy instance running in another environment. Four, single application on multi-cloud; same instance of an services runs across cloud providers to keep critical applications better positioned against denial of service attacks. Five, cloud arbitrage; workloads may be dynamically placed in environment with the most cost competitiveness and with tools like kubernetes more enterprises might arbitraging some of the applications.

In a multi-cloud environment, each cloud will have different tools and approaches. Organizations need to set and standardize policies and aggregate information for consistent governance. Some key attributes which needs to be considered are providing centralized visibility, structuring cloud accounts, standardizing access, defining and implementing tagging, implementing cloud policies, automating cloud policies, implementing cost management, standardizing images/templates, implementing patching for update processes, ensuring monitoring and uptime, standardizing audit trails etc. All the above requirements give rise to need for effective multi-cloud management platform.

Finally, a multi-cloud management platform should cover areas such as: Ability to predict, track, understand, and optimize costs across used clouds; Service management for businesses are under pressure to provide services across environments; Comprehensive visibility of multi-cloud environments is vital for prioritizing fixes, performing audits, and business continuity; Ensuring the speed and performance of applications delivered to end users from complex multi-cloud environments; Automating orchestration of data, applications, and infrastructure across different cloud environments; Aligning the activities of security, operations, and development teams to maintain security and compliance.

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)

Please key in the two words you see in the box to validate your identity as an authentic user and reduce spam.

Subscribe to this blog's feed

Follow us on