top of page

Distributed Stateful Edge Platforms

Eases the re-platforming of servers from VMs to containers.



The 48th IT Press Tour had the opportunity to meet with Partha Seetala, President of Cloud Business Unit, Rakuten Symphony. I had met with Partha several times previously as the founder and CEO of Robin.io. Rakuten Symphony acquired Robin in April 2022.


Why Distributed Stateful Edge?


It's becoming more important for compute and apps to run closer to where the data is being produced. If data is not filtered, processed, and acted on closest to where it is generated, it results in a poor user experience, costs to transfer data over the network, and poor end-to-end performance.

Data coming into cell towers, distributed point-of-sale systems, and HD video collection are driving the adoption of Kubernetes to more use cases.


Challenges Managing Distributed Stateful Edge


There are several:

  • Managing the infrastructure at tens of thousands of locations. Management includes installations, upgrades, operating systems, and platform software.

  • Deploying, healing, and managing applications on all of these servers.

  • Having a holistic view into the health and performance of infrastructure and applications at all of these locations.

  • Ensuring security at scale.

  • Managing and containing energy costs.

Distributed Stateful Edge Cloud


Symworld Cloud provides three products to provide stateful, 5G, and edge applications with a performance at scale. The solution is gaining traction for data-intensive applications with app-aware storage and data management for Kubernetes, cloud-native deployment of RANm core, and edge apps, and the management of bare metal servers, network elements, and network services at scale.


Cloud-native platform (CNP) is a fully integrated enhanced Kubernetes platform, optimized for running storage and network-intensive services with a focus on zero-touch deployment and operational automation at the edge.

CNP use cases include:

  • Deployment of stateful apps on Kubernetes

  • Deploy and manage apps at the edge

  • Common platform for hybrid and multi-cloud

  • Deploy 4G/5G RAN, core, and NFV at scale

  • Common platform to run containers and VMs

  • Automate day-2 operations

Cloud-native storage (CNS) is a high-performing cloud-native storage stack for Kubernetes with a special focus on storage and application-aware data management where volumes and applications can be manipulated as a whole as a single logical group.


Day 0 capabilities include data persistency for stateful apps, always-on availability with self-healing, and advanced data placement for stateful apps. Day 1+ capabilities include predictable and stable performance, elastic scale capacity over time, observability to enhance maintainability, and data management to ensure business continuity.


Data management capabilities include snapshots, clones, back-ups, and multi-cloud portability with simplified operations via a one-click GUI, single CLI command, and API triggers.

Orchestrator is a highly scalable infrastructure and service application orchestrator to manage bare metal servers and applications across 100,000+ servers, 10,000+ clusters, and data centers. This is accomplished with: bare metal-as-a-service for BIOS, OS, FPGA, Robin K8s install, monitor, and upgrade; network function-as-a-service for deploy, heal, monitor, scale, and upgrade; and MOPS manager for scalable and extensible management across hundreds of thousands of network elements.


Key Takeaways

The Symworld Cloud Distributed Stateful Edge Platform provides:

  • An optimized Kubernetes platform for running storage and network-intensive applications with built-in advanced storage and data management stack, advanced networking, advanced compute (CPU and GPU) for low-latency/high-throughput applications, and advanced scheduling and observability.

  • One common platform to run apps as CNFs (in containers) and VNFs (in virtual machines).

  • Automation and orchestration at three layers: infrastructure (bare metal servers), platform (Kubernetes), and application services (CNFs, VNFs).

  • Centralized infrastructure-to-application observability and policy-based closed-loop operations.

Comments


bottom of page