Fierce’s Multi-Cloud Networking 101 series takes a look into the future, using a series of progressive interviews with experts to help readers better understand the key foundational concepts and market landscape of multi-cloud networking. This emerging field aims to solve the problems associated with linking networks and applications across multiple cloud environments.
This week's interview features Dean Bogdanovic, founder and CEO of Volta Networks. The interview transcript has been lightly edited.
Fierce Telecom (FT): Are there different styles or kinds of multi-cloud networking?
Dean Bogdanovic (DB): Multi-cloud networking can be thought of as having different architectures and topologies. The architectures are public to public, public to private, and private to private. All of these are based on how you get your cloud services. The topology is based on how workloads are distributed. One of several topologies is a mirror topology where the public cloud mirrors the private cloud. Typically, one will be used for production and the other for staging and development. The private cloud is generally used for development and public for production workloads. A mesh topology is a flat network spanning multiple cloud environments with any-to-any connections among all cloud resources. Further, this can be tiered, partitioned or burst. For example, if capacity is available, production traffic could burst into the development environment. This adds some complexity but can result in better resource utilization.
FT: How long does it take to roll out a multi-cloud networking platform and what is involved in the process?
DB: Once an organization has decided on its architecture and topology then it is a fairly easy process to set up the multi-cloud network. However, it is critical that network traffic policies are established and deployed correctly. Multi-cloud hybrid environments – meaning multiple cloud vendors – can make the network traffic policies more complex. This also can be the case with network security.
FT: What benefits are associated with multi-cloud networking?
DB: Multi-cloud networking is a key enabling technology for enterprises to control their cloud computing costs. Increasingly, cloud-based compute capacity is becoming commoditized. It can be thought of as utility not unlike electricity. Innovation from the cloud computing providers spreads quickly resulting in very similar features. Enterprises can monitor their demand for compute capacity in different areas and shift workloads based on availability and the lowest cost including transit costs. This provides for much more flexibility and cost optimization.
FT: What kinds of companies and verticals are expressing the most interest in multi-cloud networking today? Do you expect this to change over the next year?
DB: Large enterprises, particularly multinational organizations, are more likely to have more complex compute requirements and traffic patterns. As a result, these organizations will benefit from more aggressive cost control. For example, they can shift compute workloads to a lower cost data center that could also reduce their transit costs. Organizations that use significant amounts of cloud computing are most likely to benefit from using multi-cloud networking in order to optimize their costs.
FT: Why is it important for enterprises to start thinking about multi-cloud networking now rather than later on in their cloud journey?
DB: Once an enterprise starts using multiple cloud providers, multi-cloud networking can provide benefits particularly cost savings. The scale of the cloud implementation will determine if it makes economic sense to actively manage in order to control costs.
FT: What is the biggest hurdle to multi-cloud networking today, and how can this be resolved?
DB: As noted, there is some complexity to multi-cloud networking in both network traffic policies and security. Enterprises would benefit from more robust automation and verification/assurance tools which are lacking today.