Commercial edge AI server for space cloud

February 12, 2021 //By Nick Flaherty
Commercial edge AI server for space cloud
The Spaceborne Computer-2 (SBC-2) set for launch to the International Space Station next week combines a commercial edge AI computing system and Intel-based HPE ProLiant server with a GPU.

Hewlett Packard Enterprise (HPE) is set to launch a commercial edge AI server to the International Space Station (ISS) that will be able to connect to Microsoft’s Azure cloud service.

The Spaceborne Computer-2 (SBC-2) combines an edge computing Edgeline Converged Edge system designed for harsh environments in the oil and gas industry with a DL360 ProLiant server based around the Intel Xeon Scalable processor running Linux with a graphics processor (GPU).

The first Spaceborne Computer was a proof of concept that commercial systems could operate in the ISS and it was used for a year. SBC-2 replaces the 120V AC supply with a direct coaction to the ISS 28V DC supply.

SBC-2 provides twice the computing performance to ingest and process data from a range of devices, including satellites and cameras, and process in real-time. A key difference is that the system has edge AI capabilities with the GPU and can connect to the Azure cloud back on Earth.

The GPU will be used to process image-intensive data requiring higher image resolution such as shots of polar ice caps on earth or medical x-rays as well as specific projects using AI and machine learning techniques. This eliminates longer latency and wait times associated with sending data to-and-from earth to tackle research and gain insights immediately for a range of projects. These include real-time monitoring of astronauts’ physiological conditions by processing X-Ray, sonograms and other medical data to speed time to diagnosis in-space.

There are hundreds of sensors that NASA and other organizations have strategically placed on the ISS and on satellites, which collect massive volumes of data that require a significant amount of bandwidth to send to earth to process. With the onboard edge AI computing, researchers can process on-board image, signal and other data related to a range of events, such as traffic trends by having a wider look at number of cars on the road and even in car parks, air quality

Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.