Managing the Complexity of Multicloud

Guest blog from Dave McCarthy, Vice President at IDC

One of the original value propositions of cloud was simplicity. Instead of managing the intricacies of on-premises infrastructure, developers and IT professionals would be able to focus on adding value to the business. In fact, many cloud providers predicted a world where on-premises datacenters would disappear entirely.

Fast forward to today and most enterprises will agree that the cloud is anything but simple. Infrastructure has become more complex and more distributed. Hybrid architectures still exist, multi-cloud deployments have become commonplace, and clouds have been extended to remote edge locations. 

In absence of simplicity, IDC predicts that by 2025, 75% of organizations will favor technology partners that can provide a consistent application deployment experience across cloud, edge, and dedicated environments.

There is no question that cloud has become the de facto platform for innovation. The ability to quickly provision resources, scale on demand, and deploy globally has changed the way enterprises build, manage, and secure applications. 

That is why even in challenging economic times, cloud services remain resilient. IDC is forecasting the overall cloud services market to reach $655.7 million in 2023 with a year-over-year growth rate of 20.1%.

IDC researched cloud deployment patterns to understand why certain architecture decisions are made. It became obvious that having a single cloud provider approach that exclusively deployed in hyperscale datacenters did not meet the needs of customers with a mix of traditional and modern applications. Our research revealed several limitations with this approach.

First, there are reasons that on-premises infrastructure still exists. The most common are:

• Latency: The distance between an endpoint and a remote datacenter introduces network latency, which is prohibitive to real-time applications.
• Cost: As more data is created outside of datacenter environments, it can be expensive to transfer and centrally store this information.
• Data sovereignty: Whether due to government regulations, industry standards, or corporate governance, more organizations must keep data local.

Second, we found several situations that motivated customers to use more than one cloud provider:

• Capabilities: Each cloud provider has unique strengths and customers are matching workloads to specific services in a provider’s portfolio.
• Resiliency: Cloud providers are not immune to outages so diversifying workloads across multiple cloud providers can improve availability. 
• Lock-in: Enterprises want the ability to switch providers if needed and to gain negotiating power by leveraging multiple providers.

While the multi-cloud approach solved several problems for customers, it introduced new operational challenges. This includes the effort needed to integrate existing systems as well as the technical expertise required to operate multiple cloud platforms. There are also concerns for how to manage data across distributed infrastructure in a way that ensures reliability, performance, and security.

This set of challenges is driving demand for solutions that offer a consistent experience in infrastructure, operations, and management regardless of where a workload is deployed. There needs to be flexibility of choice in deployment location and whether the solutions are managed directly by the customer or in conjunction with a service provider (or in many instances, multiple service providers).

It is important to note that consistency and choice are equally important for both infrastructure and applications, which often have different stakeholders. IT operations teams put an emphasis on high resiliency, focusing on the availability, security, and cost implications of critical applications. Developers and DevOps teams need self-service automation to build infrastructure as code to remove friction and accelerate innovation with rapid delivery of new features and the ability to take advantage of newer technologies such as machine learning and AI.

The ideal solution is one that can operate multiple clouds as a single entity with essential services that span life-cycle management, cost optimization, global policy enforcement, advanced threat detection, and automated developer services. The solution must address the needs of infrastructure administrators, cloud administrators, SecOps, and DevOps.

Implementing a consistent cloud platform can accelerate application delivery and cloud migration. It also can reduce cloud spend while simultaneously modernizing hybrid infrastructure.

Here is my advice for success when managing the complexity of multicloud:

• Choose the best provider and deployment location for each workload.
• Create consistency in how infrastructure, data, and applications are managed and secured.
• Implement a developer-ready platform to help accelerate the pace of innovation.