Introducing MobiledgeX Edge-Cloud 2.0

February 27th, 2020

Headshot for Sunay Tripathi

Sunay Tripathi

Chief Technology Officer & EVP of Engineering & Product

As stated here, we have released MobiledgeX Edge-Cloud R2.0, the next version of our product.  While the R1.0 release focused on the public telecom edge, R2.0 extends the capabilities into the private edge with on-premise deployments now as easy to manage as public edge.

2019 concluded with over 100 companies engaging on MobiledgeX Edge-Cloud R1.0 and initial live cloudlets working with operators in North America, Europe, and Asia.  5G has started to roll out and it is increasingly clear that enterprises cannot afford to wait to start their next wave of digital transformation, to ensure they disrupt themselves before others do it to them.  

While enabling developers to service their customers via public or telecom edge, we came across some developers and enterprises who are trying to deploy similar applications on-premise and facing many issues around orchestration, heterogeneous infrastructures and dealing with security and privacy. These enterprises are in the retail and manufacturing spaces and deal with immersive applications across multiple locations for robotics, video analytics, video security, or hospitality. On closer look, their use cases and needs were pretty similar to what we were trying to do for public edge i.e. an enterprise has many locations in the form of retail stores or manufacturing sites where they have some infrastructure that they operate. It was exactly similar to an operator running cloudlets in many different locations and managing it from one central place. These use cases and some additional requirements such as heavy demand for privacy and often the use of applications where the enterprise operates and manages the application themselves or as a managed service instead of letting the developers manage their deployment. We solved that by making the different roles on our platform even more fine-grained where a person can be part of one or more organization (either a developer or an operator organization) at the same time and the person can also have many different kinds of roles simultaneously and the owner of an enterprise can give different roles to people from within or outside his organization.

These new capabilities allowed us to support an enterprise on our platform similar to how we supported a telecom customer and have different people in their organization be responsible for different applications and have policies around where these applications can be deployed. They can now control if the application will have access to the outside world and control that access or the application will be locked down for privacy reasons etc. The interesting side effect was that enterprise operators could have people from the system integrator world take the role of manager of infrastructure and support managed services deployment and a developer could join many different enterprise organizations to support his application as a managed application or other such combinations.

BTW, some of the capabilities help our public edge use cases as well but more importantly as we go forward, we see hybrid edge capability emerging where you can start servicing an enterprise location from a telecom edge with a fixed or static application deployment (think small retail locations) while some location can have on-premise capabilities or our distributed matching engine allows main application deployment to be on-premise but support the backup/failover form the nearby telecom edge to keep on-premise deployment more cost effective.

For most solutions today we see the automated management of distributed software systems in fixed locations to be the driving need.  We see a parallel world developing, however, where devices become more aware of the edge cloud around them and the edge clouds become increasingly intelligent in managing the support of the applications requesting to be served.  

To support this world, our software investments all resonate around 3 separate areas:

  • Device side integration and controlling the device to backend path (dynamic or static) including providing the ability to monitor, debug and manage the enterprise device and application backend from a centralized console

  • Enabling a fully dynamic world of client and backend workload deployments, where declarative business policies control real-time deployment automation 

  • And as previously mentioned - a privacy-centric control plane over a distributed PaaS layer, a federated, planet-scale kubernetes

Fixed or static deployments will gradually migrate to dynamic deployments and we are designing our system to allow that migration to be seamlessly handled, by the system rather than the application and solution developers.  In the same way your mobile phone seamlessly connects to the “best” cell tower, applications will connect to the “best” cloudlet. And as applications move, and as business policies change, and as business relationships shift, and as system resources are used, all will be controlled by a defined business policy, written by the application owner.

The new generation of customers and the new generation of employees expects everything to work like the app on their phone. Simple, quick and easy.  All businesses that want to survive need to deliver on this promise in their customer-facing experiences and end to end supply chains and operations.  Physical engagement starts to be as personalized as that received online and with that direction, there is a different need to scale solutions and real-time responses that MobiledgeX Edge-Cloud R2.0 enables.

Anybody interested in reading our first release of documentation please see here. We welcome all application developers, new device makers, enterprises and telecom operators who are interested in joining our Early Access Program and are willing to help us guide the roadmap of MobiledgeX Edge-Cloud