Edge Processing & Applications in Space

Ken Obuszewski, VP of Business Development & Product, VORAGO Technologies

July 30, 2024

What is Edge Processing? Learn About the Technology & Applications in Space

What is Edge Processing?

The goal of edge computing is to move data collection, processing, and analysis from a central server (i.e., data center) to the point of data generation (i.e., sensors and devices).

Edge computing originally emerged in the late 1990s as a method of decentralizing data processing, solving congestion problems seen in the early days of the world wide web. Historically, most data processing and analysis was performed in on-premises data centers, managed by individual enterprises. The advent of public cloud computing helped to proliferate the concept of distributed computing, including edge computing.

Benefits of Edge Processing

Edge processing provides significant architectural advantages over traditional methods. Key benefits gained by processing data at the edge rather than sending it to a central server include:

• Greatly reduced latency. Edge computing reduces latency by eliminating downtime associated with data transmission to and from central servers.

• Enhanced security. Localized processing minimizes risks associated with data being transmitted to central servers.

• Reduced power consumption. Edge computing technologies have played a crucial role in enhancing sustainability by minimizing data transmission to centralized servers, leading to reduced energy consumption and improved operational efficiency.

• Performance and bandwidth optimization. Processing data where it is collected alleviates bandwidth congestion, especially in scenarios where connectivity is limited or unreliable.

These advantages ultimately result in greater flexibility and scalability and enable real-time decision-making. 

Examples of Edge Processing Applications

Over the past 20 years, advancements in embedded computing, artificial intelligence, and machine learning (AI/ML), have led to greater needs and increased use cases for edge processing. Several edge computing models now exist to enable a broad array of applications:

• Device edge: enables computing at the device or sensor level. Examples of the device edge include cars, handheld computing devices, smart appliances and smart meters.

  • The smartphone was the primary driving factor of the “device edge” – computing on individual devices – which in turn led to rapid advancement in technologies that make the device edge a reality. This included the mass deployment of low power consumption Arm CPUs. The addition of advanced IP such as embedded graphics and AI accelerators allows for real-time decision maker at the device edge.

• Gateway edge: deployed in a small office or industrial setting utilizing small, localized servers. Often used for applications that collect and aggregate data from multiple devices and sensors.

  • With reasonably performant computing hardware, and an operating system such as Linux, a gateway can be deployed which includes the ability to manage multiple devices, deploy over-the-air (OTA) updates and perform localized AI inferencing. Popular implementations include Internet of Things (IoT) and the industrial edge with specific use cases such as visual inspection and predictive maintenance.

• Enterprise edge: usually delivered by a service provider, such as a telco or co-location data center. Examples include connected vehicles and smart cities, and remote monitoring of oil and gas installations.

  • These edge deployments leverage technology similar to a traditional data center deployment, moving it closer to the source of data generation.

Edge Processing in Space

It is easy to understand the value of edge computing in the traditional IT environment, when you consider that data will need to travel hundreds of miles or more from spacecraft or satellites to reach the central data center. Now consider how critical it is for space-based applications, when the data might need to travel a million miles or more, and without the benefit of highly reliable high speed data transport.

Benefits of Edge Processing in Space

In addition to avoiding the extreme latency challenges, in-orbit processing provides several additional advantages in space, and we can relate to traditional edge processing use cases.

• We can view a rover as similar to the device edge, and edge processing provides localized decision maker capability such as navigation and autonomous operation

• Satellites are analogous to the gateway edge, where images are aggregated from multiple cameras, with localized image and compute processing, and intelligent communication with the central server (earth)

• Networks of satellites are managed in a similar manner to the enterprise edge, providing the ability to deploy optimized end-to-end computing architecture, between the earth and the satellites

Applications of Edge Processing in Space

There are many examples of edge processing use cases in space, such as:

• Satellite imagery use cases including national defense, wildfire detection, disaster management and humanitarian aid

• Planetary exploration such as the Mars Exploration Program (MEP)

• Space weather monitoring, such as identifying solar flares

The Future of Edge Computing in Space

In recent years, the Hybrid Space architecture and public/private collaborations have accelerated the adoption of edge computing in space. The partnership between NASA and SpaceX on the development of the Crew Dragon spacecraft supporting the International Space Station is a great example of what is possible in the future. This collaboration is very beneficial to the governments and to society by greatly accelerating innovation while reducing costs and time-to-market, while creating an exciting new industry valued at over one trillion dollars by the end of the decade.

How VORAGO Enables Edge Computing in Space

VORAGO has a deep history supporting space missions with radiation-hardened microcontrollers. More recently, the company announced the addition of its VA7230 edge computing microprocessor, the first Arm®-based microprocessor (MPU) with an embedded graphics processor (GPU) targeted at space applications.

The VA7230 is specifically designed to accelerate the deployment of edge processing workloads in space. Its architecture leverages a similar approach as the previously mentioned device edge computing solutions and is designed for optimized power and performance profiles.  

• 1.5 GHz Arm Cortex-A72 CPU’s and embedded graphics the product can process the most challenging edge workloads operating below 10 Watts, well below a traditional FPGA or discrete GPU solutions. 

• Multiple high-speed interfaces enable highly efficient networking. This includes a 4-port embedded time sensitive networking (TSN) switch, which enables the implementation of deterministic, reliable networks.

Achieve Your Mission with VORAGO’s Edge Computing Microprocessor

Looking to enable edge processing workloads in space? Contact VORAGO for more information, or request a quote.