Leveraging the Resiliency of the Cloud to Efficiently Operate Existing Multicast-Dependent Workloads

Air Traffic Management (ATM) has been and continues to be the cornerstone to keeping our skies safe, with sizable investments over the last two decades to upgrade and update the National Airspace System (NAS) to maintain that charter. To continue to safely operate and increase the efficiency, decision making, and situation awareness ATM is currently charting a path for the future that will transform how future technologies are developed, implemented and how data/information is shared.In existing ATM automation systems, each system independently maintains its own technological infrastructure elements (e.g., hardware, operating systems, data management tools). This construct does not allow for the system to take advantage of shared, enterprise infrastructure elements and methodologies. Therefore, an evolution from these monolithic systems to enterprise services that can leverage layered service-based architectures with horizontal integration that supports broad sharing of mission services and data across an enterprise is the future for ATM. While this large of an endeavor will need many different capabilities and technologies to support the transition to the use of Cloud has been identified as a key enabler, this includes both fully off-premise cloud-based environment and a hybrid of off-premise and on-premise cloud environments. The goal of this transition is to support scalability, availability, and shared common data flows/infrastructure, optimize operational costs, and continue to improve ATM services.The last few years Cloud Service Providers (CSPs) have been working to provide new features and services that allow software that was previously created/developed before the Cloud to be able to be migrated to the cloud. By using a Lift-and-Shift approach which still takes advantage of the elasticity and dynamic resiliency that the cloud provides. The FAA has made huge investments over the last two decades to create applications currently being used to keep the skies safe and operationally rely on technology that was cutting-edge at the time the applications were written which in some cases could be more than a decade ago. While a lot of effort is being put into planning the replacement of these applications withCloud Native versions of them, these efforts are likely to take many years to complete. At L3Harris we believe that the use of Cloud services can offer immediate benefits in support of many ATM missions but agree that investigation was needed to address the obvious questions around performance.Therefore, L3Harris has partnered with Amazon Web Services (AWS) to develop a series of Proof of Concept (POC) activities, that have allowed for a better understanding of the current capabilities of the to host existing applications. The current POC activities are exploring areas such as recasting existing capabilities/services related to surveillance in a Govcloud environment. Existing applications could be rehosted in the cloud with very little to no changes, including ones that rely exclusively on multicast traffic for communication with downstream processes as well as for load sharing. This paper will discuss activities to date, observations noted, and lessons learned such as:•AWS’ ability to natively support intra-region multicast traffic including the use of Internet Group Management Protocol (IGMP) for dynamic multicast groups.•The AWS Transit Gateway ability to efficiently handle large amount of multicast traffic.•CloudWatch Metric Filters can be used to generate metrics from well-structured logs.•AWS’ Autoscaling Groups can provide many of the benefits that are normally associated with container-based applications including the ability to scale in and out based on a set of metrics.•AWS Aurora Global Database can be used by applications that require 4-9s+ of availability.