Cloud migration, for most organizations, is considerably more complex than simply plugging into the nearest cloud platform provider. Different IT systems and workloads may require different platforms, compute power, storage and performance levels, among other things. Organizations typically address these differing requirements by using multiple cloud providers or a mix of cloud, on-premise and collocated IT solutions. The increased use of edge computing to move data and workloads closer to end users can add additional complexity to the mix.  

All of these things are part of the new challenges facing IT departments in a multicloud world, said Dominic Romeo, director of product management for TierPoint.

 “We’ve entered the age of hybrid IT and putting the right workload in the right environment necessitates that we have multiple cloud vendors,” said Romeo. “It may not always be the most convenient from an orchestration or management perspective, but it’s the right thing to do.”

There are many reasons for locating a workload in one cloud or data center vs another including performance, cost and compliance issues.  However, most workloads also must interact with each other and with applications in other environments. Connectivity, therefore, becomes a critical factor in optimizing a hybrid IT environment.

Multicloud Connectivity Challenges

In the recent webcast Connecting Core to Edge: Investigating Multicloud Deployment Use Cases, Romeo and Will Pittman, head of Eastern sales for Megaport, a TierPoint partner and network as a service provider, discussed the challenges involved in connecting multiple cloud and data center environments.  They pointed to several key issues, including:

  • Networking costs. While the internet provides connectivity, it doesn’t provide a consistently reliable connection, so IT departments routinely turn to VPNs or other types of dedicated connections, all of which can be expensive and take time to implement. Investing in point-to-point circuits or VPN connections can become even more burdensome when connecting multiple environments.
  • With network connections between multiple platforms, end user locations, edge data centers and hyperscale cloud providers, creating a networking map that optimally serves every need is a complex challenge. Latency can become a serious problem, especially as edge computing takes off.
  • Future migrations. As companies expand or change business models, their IT systems also change. IT departments may have to move workloads and storage to other cloud or on-premise environments or add network connections. These changes should not require a complete overhaul of an IT environment.
  • Most organizations must comply with one or more government data privacy and security laws, both U.S. and European, as well as many state laws. When data and workloads are in a highly distributed environment, maintaining a consistent security perimeter around the various enterprise systems becomes difficult.

Also read - 2019 Cloud Trends: The Rise of Cloud Connectivity

The Solution to the Hybrid Cloud Issues

To address these challenges, Pittman and Romeo described two important infrastructure solutions: Network as a Service (NaaS) and software-defined networks (SDN).

Network as a Service

A NaaS provider as the name suggests, offer networking functionality on a subscription basis. Distributed cloud and hybrid environments can benefit greatly from a centralized networking infrastructure – something that is nearly impossible for most organizations to build on their own. However, a NaaS or a managed services provider (MSP) with a geographically distributed network of data centers can serve as a middleman between disparate cloud providers and data centers. A NaaS or MSP that has partnerships with major cloud platform providers (i.e. Google, Amazon and Azure) as well as the right geographic coverage can serve as the network glue in hybrid and multicloud environments.

A provider with overlapping regional data center networks can connect across wide distances as well as provide custom connections, such as between edge data centers or for local, latency-sensitive situations. Many-- such as TierPoint--also offer colocation facilities, disaster recovery and backup services and security services to help take the management burden off of IT departments.

Software-Defined Network (SDN)

A second component that can help to reduce the cost and complexity of hybrid connectivity is a software-defined network (SDN) such as Megaport, that can enable greater flexibility and scalability in connections.  An SDN programmatically describes network configurations and can be dynamic and flexible, to change as network performance requires. According to a Network World article, SDNs “separate the network control plane from the forwarding plane to enable more automated provisioning and policy-based management of network resources.”

Finally, many MSPs routinely offer advice on the best approaches to a variety of networking, security, workload placement and other hybrid IT challenges. With experts in many different specialties, an MSP can provide a wealth of information and suggestions.

“Achieving multicloud can be really simplified by having the right connectivity,” said Romeo. “But that can take having someone sit down with you and discuss a solution-based approach.”

Learn more about Multicloud Connectivity

Want to see how NaaS and SDN can solve specific connectivity issues? Watch Connecting Core to Edge: Investigating Multicloud Deployment Use Cases and learn how two companies–an energy company and an ecommerce company--were able to leverage TierPoint’s national network of data centers and the Megaport platform to solve their difficult network connectivity challenges.

The Strategic Guide to Cloud Computing - read now

Subscribe to the TierPoint blog We'll send you a link to new blog posts whenever we publish, usually once a week.