A New Option for Arm-Based Edge Computing: The V-Raptor Edge

A first look at the newly announced V-Raptor Edge computing device.

It's a new decade and it's safe to say that cloud computing is the norm. Despite the fact that we all use our own personal computers, we store and work off centralized servers with services such as Google's suite of applications (Drive, Docs, Sheets, etc) and Apple's iCloud (Photos, Notes, Pages, etc.). Most of our household devices also work from a cloud service versus local storage and processing. Amazon Alexa, AppleTV, Nest cameras, all stream and store data from their respective cloud services.

While working and streaming from the cloud is ideal for an endless list of reasons, the large amount of constant network traffic means that latency can become a potentially crippling issue. The cloud also has to physically exist on a server somewhere in the world, meaning that the companies hosting those cloud services like IBM, Google, Amazon, and Apple are paying to maintain large servers as well as ensure they are expanding fast enough to meet user demand.

To address these two downfalls of cloud computing, the concept of bringing computation capabilities and data storage closer to the end user was developed. Starting back in the late '90s to make video content a viable option via web access, content delivery networks were developed which consisted of distributed proxy servers between the main server and the end user to allow for a physically shorter paths for the video to be streamed across. These content delivery networks have evolved since then in that the proxy servers have been replaced with server nodes possessing the computing power that allows them host applications and store data themselves.

This server structure is referred to as edge computing and has been pivotal in reducing latency in networks that would otherwise be useless for latency sensitive tasks. It also saves bandwidth across the network as a whole and offloads some of the storage space requirements on the central cloud servers. These server nodes are referred to as edge nodes since they physically exists at the "edge" of the network. These edge nodes deploy virtual mechanisms such as virtual machines to be able to provide the services to run host applications and other edge specific tasks such as IoT management and security.

High level example diagram of edge computing (📷: "Edge Computing" Wikipedia)

The edge brings the cloud to the user, and with the popularity of IoT devices in the hobbyist community that are connected to cloud services it's only natural that an interest for developing dedicated edge computing devices has come about within the hobbyist community.

Large companies like AT&T and Verizon offer edge computing hardware solutions for corporations looking for large scale deployments in places such as sports stadiums with equally scaled price tags. This is not exactly a viable solution for the average hackster hobbyist. However, this past Friday a small Korean company XSLAB Inc., released the initial hardware and software specifications of their V-Raptor Edge computing device.

The V-Raptor hosts hardware specifications that are quite impressive:

  • 64bit Arm Cortex-A72 dual cores and Arm Cortex-A53 quad cores (1.8GHz)
  • 4GB of LPDDR4
  • Micro SD (32GB) or eMMC (16GB)
  • 1GB/s Ethernet
  • Wi-Fi/Bluetooth 5.0 combo
  • USB 3.0 (x2), USB 2.0 (x2), USB Type-C (DP display)
  • Graphics support to output to a single 4K display or dual full HD displays (HDMI, USB Type-C)

The V-Raptor edge runs a customized image of Xbuntu 18.04 with a UEFI that claims to be complaint with Arm's Embedded Base Boot Requirements (Arm's latest effort to standardize the boot process of embedded systems). As expected, the V-Raptor Edge's custom Xbuntu image focuses on system settings for various types of remote configurations for edge computing purposes. For example, the web browser in the V-Raptor Edge's OS has been customized to specifically support accessing remote consoles for virtual machines.

The option for firmware updates is also listed as one the the V-Raptor Edge's software features, which I'm guessing could mean the Xbuntu image will also support over-the-air firmware updates. This is just an educated guess based on the other features of the V-Raptor Edge, as the only thing the creators of V-Raptor Edge has explicitly stated is "Firmware Update supports".

The V-Raptor Edge will ship with a VDI client installed on it from the Seoul-based telecommunications company, SK Broadband. This VDI client will allow for a virtualized desktop environment of Windows 10 to be run on the V-Raptor Edge. From what I can tell, it seems that this VDI client is a subscription-based service. However, since this service from SK Broadband is only about a year old I struggled to find a clear way to sign up for the Cloud PC service being an international customer, so hopefully V-Raptor will ship with a trail license already installed or directions for how to obtain a license.

In a YouTube playlist of short videos, the V-Raptor Edge is shown being put through its paces connecting to Wi-Fi, Bluetooth, dual HD monitors, and running Cloud PC to connect to a Windows 10 virtual desktop.

V-Raptor Edge (📷: @FunfunYoo on Twitter)

Release is currently expected to be in late April or early May of this year according to V-Raptor’s CEO for a price point of $300 not including tax or the Cloud PC VDI service from SK Broadband. Online sales will eventually be available, although V-Raptor’s CEO also mentioned that partner businesses selling the V-Raptor Edge are currently the plan with that list of partner business to be determined at the moment.

A brief look at the inner workings of the V-Raptor Edge with its lovely purple PCB.

V-Raptor's CEO posts regular updates on the V-Raptor Edge project on his Twitter account and is very interactive to inquires about the product. I'll be keeping an eye out there for when the V-Raptor Edge becomes available on the market!

whitney-knitter

All thoughts/opinions are my own and do not reflect those of any company/entity I currently/previously associate with.

Latest Articles