Cloud-Network Convergence: Double-Edged Sword for Carriers, but an Opportunity Emerges in the Edge

M. Rogers

Summary Bullets:

  • Hyperscale cloud providers continue to invest in expansion of networking solutions and network adjacent partnerships with SD-WAN vendors and data center providers.
  • As hyperscale networking solutions become more advanced and the cloud edge moves closer to end users, some carriers may be crowded out of the cloud-network convergence market.

Hyperscale cloud providers and global network carriers have been building stronger relationships and deeper integrations around WAN and networking solutions. The partnerships are a natural fit as more enterprises move business-critical workloads to cloud platforms. These workloads require high-bandwidth, high-throughput, low-latency, secure, and reliable connections from virtually any location these days to connect to the major cloud platforms.

Leading telcos continue to directly connect their networks into hyperscaler cloud PoPs and many are automating this process through APIs with carrier-neutral providers like Equinix. Telcos are also working with SD networking vendors or offering their own in-house solutions to provide application acceleration, bandwidth bursting, and security to support the enterprise transition to the cloud. COVID-19 is also accelerating workload migration to the cloud and changing the requirements of networks along the way.

However, leading cloud providers like Azure, Google Cloud Platform, and Alibaba Cloud are making moves that may begin to disrupt these symbiotic relationships. Increasingly, the hyperscalers look like global carriers, own their own undersea cables, and leverage their own internal networks to provide solutions to end-customer enterprises. Many rival some of the Tier 1 carriers in terms of total fiber in the ground and CapEx spending.

Furthermore, the hyperscalers are opening up their internal networks and providing traffic optimization and application acceleration services. Examples include AWS Global Accelerator, Microsoft Virtual WAN, and/or Google’s Cloud VPN. Most hyperscalers will continue to add SD-WAN vendors to their platforms, and some will support orchestration in cloud-native environments. While enterprise still requires last-mile connectivity, the cloud edge continues to extend further towards end users. All major hyperscalers have intentions to add more edge locations, meaning connecting into the hyperscaler backbones will require fewer hops and be easier to manage.

Hyperscalers cannot replace network providers and will not look to acquire local telecom licenses. Most enterprise are pursuing a hybrid and multi-cloud approach, meaning a mix of on-premises, private, and public cloud. This means they need connectivity across their own footprints, into private data centers as well as into the various hyperscaler networks. However, as cloud and networks converge, carriers will begin to face a familiar problem: cloud becomes a reseller market and the network gets further commoditized.

With deeper network and cloud integration, carriers must drive differentiation through AI-based platforms which offer value-add such as agility and automation. Working with the hyper-scale community means the inevitability of working at an accelerated pace and never having clear-cut service boundaries. The opportunity for carriers is in consulting and services integration. Each market is different, and global offers at scale need local context. There is also an opportunity with 5G services and new use cases around edge compute. This will be disruptive to centralized cloud and mean infrastructure will be even more local and highly distributed to support the next generation of workloads like tele-haptics and industrial automation. The lowest-latency solutions can only be achieved through carrier colocation. This could mean while global WAN markets wane, the last-mile increases in significance and complexity.

What do you think?

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.