VANCOUVER, British Columbia — At last year’s OpenStack Summit 2017 in Boston, Beth Cohen, a distinguished member of the technical staff at Verizon, introduced an OpenStack-at-the-edge use case. And today at the OpenStack Summit 2018, Cohen said the community has spent much of the last year arguing about the definition of “the edge.”
“At an open developers’ meetup in San Francisco in the Fall of 2017, we had 200 people arguing about: what is edge computing?” said Cohen.
Ildiko Vansca, an ecosystem technical lead with the OpenStack Foundation, said the arguing continued at the OpenStack Summit in Sydney, Australia, in November 2017. So, a group of OpenStack members decided to write a white paper titled “Cloud Edge Computing Beyond the Data Center.”
Vansca and Cohen said the white paper was a collaborative effort with people across many industries, companies, and open source communities. It’s become part of the edge computing initiative at the OpenStack Foundation. “It defines not only edge computing but more importantly some high-level use cases,” said Cohen. “5G is not the only use case.” Other use cases will be network functions virtualization (NFV) applications at telcos, internet of things (IoT), artificial reality (AR), and gaming, to name a few.
“The defining need that drives cloud edge computing is the need for service delivery to be closer to users or end-point data sources,” states the white paper. “Edge computing environments will work in conjunction with core capacity but aim to deliver an improved end user experience without putting unreasonable demands on connectivity to the core.”
As far as defining edge computing, the paper says, “For our purposes, the most mature view of edge computing is that it is offering application developers and service providers cloud computing capabilities, as well as an IT service environment at the edge of a network. The aim is to deliver compute, storage, and bandwidth much closer to data inputs and/or end users.”
At a briefing session with media today, Cohen said it is still ambiguous what constitutes edge devices. “They can be very small,” she said. “But we got a little fuzzy that IoT devices probably aren’t edge. But they could be as small as a cell phone.”
Many Groups Work on Edge
The OpenStack Foundation is not the only group working on edge computing problems. “There are more than 20 groups trying to figure out edge computing,” said Vansca, who mentioned MEF, ONF, ETSI, and The Linux Foundation, along with the OpenStack Foundation, to name a few. “We are collaborating with the adjacent communities,” she added.
While the above-named groups are all involved with communications networking, Cohen said edge computing is definitely not limited to telco networks. “In my opinion, SDN and 5G are table stakes,” she said. “The other use cases will be following on after those use cases. The use case that’s the most commercially viable today is the telco use case.”
According to the white paper, edge computing will provide two key improvements: reduced latency and a reduction in bandwidth limitations.
Vansca said the OpenStack Foundation is a good place to work on edge computing problems because OpenStack provides modular components. “We are trying to use OpenStack components that are relevant, such as Keystone or Glance,” she said. “We stepped away from this big monolithic software that OpenStack used to be. In the edge discussions, we are looking at which components we need.”
The white paper identifies some of the requirements that edge computing will need. These include:
- A virtual machine/container/bare-metal manager.
- An image manager in charge of template files such as virtual machine and container images.
- A network manager in charge of providing connectivity to the infrastructure.
- A storage manager, providing storage services to edge applications.
- Administrative tools, providing user interfaces to operate and use the dispersed infrastructure.
- A way to address storage latency over WAN connections.
- Reinforced security at the edge to monitor the physical and application integrity of each site.
- Monitoring of resource utilization across all nodes simultaneously.
- Orchestration tools that manage and coordinate many edge sites and workloads, potentially leading toward a peering control plane or self-organizing edge.
- Orchestration of a federation of edge platforms.
- Load balancing across geographically distributed hardware.