Language Selection

English French German Italian Portuguese Spanish

Servers/Back End: Kubeflow, Kubernetes and EdgeX Foundry

Filed under
Server
  • Designing an open source machine learning platform for autonomous vehicles

    Self-driving cars are one of the most notable technology breakthroughs of recent years. The progress that has been made from the DARPA challenges in the early 2000s to Waymo’s commercial tests is astounding. Despite this rapid progress, much still needs to be done to reach full autonomy without humans in the loop – an objective also referred to as SAE Level 5. Infrastructure is one of the gaps that need to be bridged to achieve full autonomy.

    Embedding the full compute power needed to fully automatise vehicles may prove challenging. On the other hand, relying on the cloud at scale would pose latency and bandwidth issues. Therefore, vehicle autonomy is a case for edge computing. But, how to distribute and orchestrate AI workloads, data storage, and networking at the edge for such a safety-critical application? We propose an open-source architecture that will address these questions.

    [...]

    In order to implement an open-source machine learning platform for autonomous vehicles, data scientists can use Kubeflow: the machine learning toolkit for Kubernetes. The Kubeflow project is dedicated to making deployments of machine learning workflows simple, portable and scalable. It consists of various open-source projects which can be integrated to work together. This includes Jupyter notebooks and the TensorFlow ecosystem. However, since the Kubeflow project is growing very fast, its support is soon going to expand over other open-source projects, such as PyTorch, MXNet, Chainer, and more.

    Kubeflow allows data scientists to utilize all base machine learning algorithms. This includes regression algorithms, pattern recognition algorithms, clustering and decision making algorithms. With Kubeflow data scientists can easily implement tasks which are essential for autonomous vehicles. These tasks include object detection, identification, recognition, classification, and localisation.

  • Kubernetes communication, SRE struggles, and more industry trends

    As part of my role as a senior product marketing manager at an enterprise software company with an open source development model, I publish a regular update about open source community, market, and industry trends for product marketers, managers, and other influencers. Here are five of my and their favorite articles from that update.

  • Introducing a Tech Preview of Containerized Ceph on Kubernetes

    We have been hard at work to bring a containerized version of Ceph to Kubernetes, and we are very excited to announce that we are releasing a technical preview of our project to run SUSE Enterprise Storage (powered by Ceph) on SUSE CaaS Platform (powered by Kubernetes). We leverage the most modern, powerful application management framework to make Ceph lifecycle management easier, and we provide an easy way for SUSE CaaS Platform users to get Kubernetes-native persistent storage for their Kubernetes cluster backed by enterprise-grade software-defined storage.

    [...]

    The good news is that work on Rook and Ceph-Rook integration is a concentrated effort upstream. There are many eyes—and many fingers—working to make Ceph better on Kubernetes. We at SUSE are in a good position to make sure that Ceph and Rook work upstream will meet the unique needs of our customers, and we are thrilled that our customers and their needs are able to make upstream better.

  • Making The IoT More Open: A Common Framework For IoT Edge Computing With EdgeX Foundry

    The internet of things (IoT) is a diverse space, but it’s also fragmented by design, whether it’s consumer IoT or industrial IoT. In 2015, Dell started working on a project called Project Fuse to weave together the diverse and fragmented world of IoT. The idea was to build the right architecture for IoT and edge computing.

    The team working on the project quickly realized that they needed to extend the cloud-native principles — things like microservice-based architectures and platform independence — as close as possible to the device edge so that there would be more flexibility in how solutions are devised. In order to succeed, the project needed to be vendor-neutral, interoperable and open.

More in Tux Machines

System76 announce more Linux laptop models get open source firmware

System76, the Pop!_OS Linux distribution maker and hardware vendor for laptops, desktops and servers has announced another open source firmware push. Announced on Twitter, the official account mentioned that the Gazelle and the Adder WS models are joining the ranks and if you own them you can switch them over to their open source firmware right now. Not only that, their Darter Pro laptop model will get an automatic update to move it over to their open source EC (Embedded Controller) firmware. You can see here how to move your devices over. Founder and CEO of System76, Carl Richell, announced on Twitter: "I highly recommend switching to System76 Open Firmware if you have one of these models. Not only is your firmware largely liberated from proprietary code, your laptop will boot faster and you'll start seeing integrations between hardware (through this firmware) and Pop!_OS.". Read more

Games: Songs of Syx, Super Mario 64 and Stadia

  • Fantasy grand strategy city-builder Songs of Syx is out in Early Access

    As one of the most promising indie games this year, Songs of Syx has properly entered Early Access on Steam so you can try your hand at city-building with a grand strategy theme. It's hard to fully grasp the scope of it right now but it's big. You start off as an insignificant colony and build, scheme, and fight your way towards a metropolis and empire. Funded with a successful Kickstarter campaign that ended back in May 2020 with about £23K from over 800 backers. As they said it would, Linux support is wired up and ready right away.

  • You can now play Super Mario 64 natively on Android, no emulator required

    Forget Super Mario 3D All-Stars. You can now play Super Mario 64 on your Android phone without the need for an emulator. The game now has an unofficial native Android port thanks to XDA member VDavid003. In the summer of 2019, Super Mario 64 was successfully decompiled and translated into human-readable C code by a team known simply as a “group of talented individuals.” This code has been available on GitHub for a little over a year at this point, and VDavid003 took this code to create the tools needed to compile the game for Android. [...] VDavid003 has created a repo containing everything needed to compile the game on a Windows or Linux PC, which can then be sideloaded as an APK to an Android device.

  • Stadia pushing more indie games with Stadia Makers, PUBG dropping keyboard and mouse queue

    Google has announced another wave of indie games are confirmed for Stadia, their game streaming service powered by Linux and Vulkan. This is all part of the previously announced Stadia Makers program back in March during their Google for Games Keynote, where Google will directly support smaller teams using the Unity game engine to bring them to Stadia. As a result, another 7 have been announced to release at various dates.

TechNexion Unveils EDM and AXON SoM’s Powered by NXP i.MX8M Plus SoC

The company offers standard support for Ubuntu 20.04 LTS, Linux-built Yocto Project, and Android 10, as well as extended support for FreeRTOS. If it feels like you’ve seen EDM-G-IMX8M-PLUS module before it’s because it should be the one found in the upcoming Wandboard 8MPLUS SBC. There’s will be other development kits based on existing AXON/EDM baseboards including AXON-PI Raspberry Pi-like starter board, or the full-featured AXON-WIZARD and EDM-WIZARD evaluation boards. Marcel vandenHeuvel, TechNexion’s CEO, gives an overview of the AXON i.MX8M Plus modules and baseboard, and shows a Yocto 3.0 Linux demo with dual displays. Read more

Security: AppArmor and SELinux, Linux Security Features and Zero Trust Security Model

  • Technologies for container isolation: A comparison of AppArmor and SELinux

    I researched how containers, virtual machines (VMs), and processes, in general, are separated by different technologies—namely, AppArmor and SELinux. My goal was to compare these solutions for isolation/separation capabilities in the cloud world. Just as a reminder, Red Hat Enterprise Linux uses SELinux technology to separate processes, containers, and VMs. OpenShift also uses this technology. The first option is an isolation technology called AppArmor, which is a very similar technology to SELinux. However, it is not label-based. AppArmor security profiles, which are equivalent to SELinux security policies, look more user-friendly, but that’s because AppArmor is less complicated and controls fewer operations.

  • Kees Cook: security things in Linux v5.7

    Linux v5.7 was released at the end of May. [...] After Silvio Cesare observed some weaknesses in the implementation of CONFIG_SLAB_FREELIST_HARDENED‘s freelist pointer content obfuscation, I improved their bit diffusion, which makes attacks require significantly more memory content exposures to defeat the obfuscation. As part of the conversation, Vitaly Nikolenko pointed out that the freelist pointer’s location made it relatively easy to target too (for either disclosures or overwrites), so I moved it away from the edge of the slab, making it harder to reach through small-sized overflows (which usually target the freelist pointer). As it turns out, there were a few assumptions in the kernel about the location of the freelist pointer, which had to also get cleaned up.

  • Zero Trust Security Model

    The Zero Trust Network, also called Zero Trust Architecture, is a model that was developed in 2010 by the principal analyst John Kindervag. A zero-trust security system helps to protect the enterprise system and improves cybersecurity. [...] The Zero Trust approach depends upon modern technologies and methods to achieve the target of securing an organization. The Zero Trust Model calls for businesses to manipulate micro-segmentation and granular perimeter execution based on users, their whereabouts, and other data or information, to find out whether to believe a user, machine, or application that is trying to seek access to a specific part of the enterprise or organization. Zero Trust also takes care of all other policies, for example, giving users the least access they require to complete the task they want to complete. Creating a Zero Trust environment is not only about putting into practice the separate singular technology associations; it is also about using these and other technologies to impose the idea that no one and nothing should have access until they have proven that they should be trusted. Of course, organizations know that creating a Zero Trust Security Model is not an overnight achievement. Because it is not easy to achieve something so complex in a single night, it can take years to find the most secure, ideal system possible. Many companies are shifting to the cloud security system. These systems have the best options to go to Zero Trust. Now is the time to be ready for a Zero Trust transition. All organizations, either large or small, or should have Zero Trust security systems for their data safety.