Edge cloud computing is predicted to become one of the fastest growing technological sectors.
A significant part of the predicted massive data growth is associated with data generated at the edge of the network due to use cases such as IoT, automotive, AR, gaming, and 5G. IoT devices alone are expected to create over 90 zettabytes of data according to the International Data Corporation (IDC) Data Age 2025 report.
The Emergence of Edge
The emergence of AI in real time applications and the need for bare metal infrastructure near the users also creates the need for edge clouds. According to Gartner, 75% of the data will be processed at the edge by 2020, in contrast to 91% of the data that is generated and processed today at the cloud data centers at the core of the networks. Data will need to be processed and stored at the edge due to economics, connectivity, and the requirements of low latency applications.
Although there are some similarities between edge clouds and core clouds, there are some striking differences between them. Users expectations are different. Users demand lower response time from edge clouds that core clouds cannot meet due to their geographical distances from edge devices. Other differences are relatively few core cloud data centers versus many small edge clouds, geographic locations, and network connectivity.
Considerations for Edge
Discussions of edge clouds frequently focus on computing and networking considerations, but edge cloud storage is just as important, if not more so.
Edge cloud storage infrastructure must provide:
– Low latency response times
– Ability to run on a variety of servers that fit the edge location’s constraints in terms of size, density, rigidity, power, and form factor
– Ability to consolidate workloads
– Agility in adding new applications and adopting new technologies
– Fully automated and reliable hands-off operation
– Low total cost of ownership (TCO) when compared with the core public clouds
How to Optimize Storage for Edge Cloud
To meet these edge cloud storage needs, Lightbits™ LightOS® is the optimal storage disaggregation solution for the edge cloud. It is based on standard TCP/IP networking, provides extremely low latency, and high performance even with lower grade NVMe drives. The disaggregation of storage from compute enables stateless application servers and allows you to efficiently, quickly, and independently scale compute and storage according to application needs.
LightOS also supports inline compression and erasure coding and reduces TCO by enabling edge clouds to leverage NVMe drives instead of HDDs that cannot support many of the edge applications needs. Plus, it is a software-defined disaggregation solution that can run on any edge server.
To learn more about edge clouds, edge cloud storage, and LightOS, download our new white paper: Storage at the Edge.
Additional Resources
NVMe over TCP
Kubernetes Persistent Storage
NVMe Over TCP and Edge Storage
Ceph Storage
Disaggregated Storage
