Month: September 2026

CANcloud OTA Batch Manager: Update Your Whole CANedge Fleet at Once

CSS Electronics’ new CANcloud OTA batch manager lets you push configuration, firmware, password and TLS certificate updates across your entire CANedge fleet in a single operation. Until now, every over-the-air (OTA) update in CANcloud — its free browser tool for managing CANedge2 and CANedge3 CAN bus data loggers — had to be pushed to one device at a time. This overview is based on CSS Electronics’ September 2026 newsletter and its CANcloud documentation. Pertech supplies CSS Electronics’ CANedge data loggers and CANmod modules in Israel, so we’ve summarized what’s changed and where it matters for fleet operators.

CANedge3 CAN bus data logger with SD card, GPS/IMU antenna and 3G/4G connectivity
The CANedge3: one of the CAN bus data loggers CANcloud’s new OTA batch manager can now update at fleet scale. Photo: CSS Electronics.

What is CANcloud?

CANcloud is CSS Electronics’ 100% free, open-source browser tool (published on GitHub) for managing CANedge2 and CANedge3 loggers and the log files they upload to your own S3-compatible server — AWS, Google Cloud, Azure or a self-hosted MinIO instance. From CANcloud you can browse and configure connected devices, monitor firmware and storage status, and push over-the-air (OTA) updates without installing anything beyond a browser.

From one device at a time to a full fleet in one operation

Previously, every OTA action in CANcloud — a configuration change, a firmware upgrade, a password reset, a certificate swap — had to be applied to one logger at a time. For a fleet running dozens or hundreds of CANedge devices, that made routine maintenance slow. The new batch manager keeps the same underlying update mechanisms but lets you target a whole cohort of devices, selected from a searchable device table, and run the update as a single batch.

CapabilityWhat it does
Partial config updatesChange just the settings that need to change — e.g. split size or filters — in the config editor on one device, then transfer only that edit to the batch. It’s merged on top of each device’s own Configuration File, so every other setting is left as-is.
Fleet firmware updatesUpload one firmware.bin and push it to every compatible device. If the new firmware needs a newer Configuration File schema, CANcloud migrates and writes each device’s config automatically before the firmware itself.
Password rotationEnter new passwords once, in plain text, in the config editor, then enable “Encrypt passwords” for the batch. Each password is encrypted individually with that device’s own public key, so no cleartext credential is ever deployed.
TLS certificate rolloutDeploy a new certs_server.p7b bundle fleet-wide. Include both the old and new certificate in the bundle during a transition so devices stay connected before and after the switch.
Per-device validationEvery device is checked against its own Rule Schema before it can be selected, and re-validated against fresh data immediately before the update is written. Incompatible devices are excluded automatically, with the reason shown.
Searchable device tableSearch and sort the fleet by device ID, meta name, device type, firmware version, last heartbeat, config sync state or status — e.g. to isolate every device still on an old firmware build and select the whole cohort in one click.
Diff view and live statusReview a per-device diff of exactly what will change before submitting a batch, then follow each device’s update status as the run progresses.

Real-world example: AKSA’s E-COOLPAC fleet on CANedge3

AKSA Würenlos AG builds the E-COOLPAC, a fully electric mobile battery system that powers transport refrigeration units. Its customers need CAN bus data from the E-COOLPAC for monitoring, route planning and diagnostics. AKSA reads that data — from inverters, an on-board charger, a battery management system and several gateways talking CANopen and J1939 over a mobile PLC-based CAN bus — with a CANedge3. The logger stores the raw data to SD and uploads it over LTE to AKSA’s AWS S3 server whenever coverage is available, where it’s automatically DBC-decoded into physical values for AKSA’s own MYPAC telematics dashboard, giving dealers, technicians and customers traceability across each E-COOLPAC unit’s lifecycle.

“The CANedge3 is easy to integrate for transferring data to the cloud,” says Benjamin Ambiel, Product Manager at AKSA. With the data online, AKSA’s customers can plan truck routes around battery state of charge and schedule maintenance before a failure happens, while AKSA’s own service technicians can read machine data remotely rather than waiting for an on-site visit.

CANedge data loggers through Pertech: Pertech supplies CSS Electronics’ CANedge CAN bus data loggers and CANmod modules in Israel, alongside our other test, measurement and firmware tools. If you’re planning a fleet rollout, a firmware or certificate migration, or an S3 server setup, tell us about your fleet and we’ll help you plan the update.

Frequently asked questions

Do I still need to update CANedge devices one at a time?

No — that’s what the batch manager changes. Config, firmware, password and certificate updates in CANcloud can now be selected from the device table and pushed to a whole cohort of devices in a single run, with each device still individually validated and updated behind the scenes.

Is CANcloud free to use?

Yes. CANcloud is CSS Electronics’ free, open-source browser tool for managing CANedge2/CANedge3 devices and the data they upload to your own S3-compatible server. There’s no separate license fee for the OTA batch manager.

Can I roll out a new TLS certificate without losing connectivity to devices already in the field?

Yes, as long as the certs_server.p7b bundle used for the transition batch includes both the old and new certificate. That lets each device authenticate with either one until you retire the old certificate in a follow-up update.

Further reading: CSS Electronics’ CANedge documentation, the CANcloud project on GitHub, and the full AKSA E-COOLPAC case study.


Start Building with the Latest Arm Tech: AI Portal, Mali G2-Ultra NX and Performix

At Arm Create Shanghai, Arm followed its keynote with three developer launches: the Arm AI Portal for finding and deploying Arm-optimized AI models, the Mali G2-Ultra NX GPU with Arm Neural Technology for mobile graphics, and Arm Performix for profiling workloads on Arm-based cloud servers. This guide summarizes what each one is, who it is for and where to start, based on Arm’s own announcements dated 8 September 2026. Pertech is an Arm representative in Israel, so we also point out which of the three matters for embedded and edge teams, and which does not.

Infographic: three ways to start building on Arm - Arm AI Portal, Mali G2-Ultra NX with Arm Neural Technology, and Arm Performix
The three developer entry points Arm announced after the Arm Create Shanghai keynote. Source: Arm Developer and Arm Newsroom.

The three Arm announcements at a glance

AnnouncementWhat it isBest forWhere to start
Arm AI PortalArm-optimized AI models, runtimes, sample code and tools, from cloud to edgeTeams deploying AI on Arm CPUs, GPUs and NPUsdeveloper.arm.com/ai
Mali G2-Ultra NX + Arm Neural TechnologyThe first AI-native Mali GPU, plus neural upscaling and denoising techniquesMobile game and graphics developersNeural Graphics Development Kit
Arm PerformixA performance analysis toolkit for Arm Neoverse workloadsCloud, platform and runtime engineersArm Performix page

Arm AI Portal: Arm-optimized AI models and agent-ready resources

Arm AI Portal is a single place to find, evaluate, optimize and deploy machine learning models on Arm. Arm positions it for the more than 22 million developers building on Arm, and for their AI agents. At launch it includes pre-optimized models such as Alibaba Qwen, Google Gemma and Ultralytics YOLO, with supported runtimes including ExecuTorch, LiteRT and ONNX Runtime. Arm-optimized models are also available through Hugging Face.

The portal organizes models by task, which makes it easier to shortlist options for a specific product:

  • Generative AI: text generation, text-to-image and image-to-image.
  • Computer vision: image classification, object detection, segmentation and depth estimation.
  • Audio and NLP: automatic speech recognition.
  • Neural graphics: frame rate upscaling and super sampling.

Target hardware spans cloud CPUs, mobile CPUs and GPUs, embedded NPUs and edge Linux systems, and Arm lists ecosystem partners including ExecuTorch, Raspberry Pi, Ultralytics and vivo. Arm reports that Qwen3-TTS ran more than 4x faster on a vivo X300 smartphone, and that Ultralytics YOLO26n gained more than 40% performance using optimized precision. These are Arm’s own figures, so validate them on your target device.

For teams that use AI coding assistants, the portal’s resources can also be exposed to coding agents through MCP. Arm describes this agent-ready access as early access ahead of general release. Details are in Arm’s launch announcement.

Mali G2-Ultra NX and Arm Neural Technology: desktop-class graphics on mobile

The Mali G2-Ultra NX is described by Arm as the first AI-native Mali GPU. Rather than treating AI as a separate block, it places neural accelerators inside the shader cores, so neural graphics workloads run alongside conventional graphics and compute and reuse the GPU’s memory system and caches. According to Arm’s technical blog, the GPU also brings:

  • A redesigned execution engine, the largest Mali instruction set update in seven generations, with up to 2x more registers per warp.
  • Third-generation ray tracing, with up to 13% lower DRAM traffic on leading ray tracing benchmarks and hardware support for Opacity Micromaps. In one Arm demo, Opacity Micromaps raised frame rates by 30% and cut the ray tracing workload by up to 70%.
  • Higher baseline performance: up to 24% higher benchmark performance and 14% higher non-AI gaming performance than Mali G1-Ultra.

The software side is Arm Neural Technology: Neural Super Sampling (NSS) reconstructs higher-resolution images from lower-resolution renders, Neural Frame Rate Upscaling (NFRU) generates intermediate frames for smoother gameplay, and Neural Super Sampling and Denoising (NSSD) cleans up ray-traced scenes. Arm’s Neural Graphics Development Kit packages Unreal Engine plugins, tools, documentation, sample code, Vulkan ML extensions, pre-trained models on Hugging Face and a Neural Graphics Model Gym for training custom models. NFRU is publicly available, and an Early Access Program covers the rest ahead of release.

Arm’s performance claims come from its own testing between March and August 2026, and actual results may vary. Mali is silicon IP that chip makers build into their own SoCs, so it reaches developers through the phones and devices that adopt it.

Arm Performix: find performance bottlenecks on Arm cloud infrastructure

Arm Performix profiles running applications, collects system-level and hardware-level metrics, and turns them into guided insights. It is aimed at platform engineers, core library developers, runtime and compiler engineers, and cloud backend developers. Key capabilities listed by Arm include:

  • Finding hotspots across compute and memory, with CPU microarchitecture analysis and clear visualizations.
  • Detecting NUMA and heterogeneous compute interactions that hurt performance.
  • CI/CD integration for performance regression testing.
  • Dynamic insights through AI agent integration, compatible with the Arm MCP Server.

Performix targets Arm Neoverse-based Linux systems, including AWS Graviton, Microsoft Cobalt and Google Axion, in cloud or on-premises deployments. Arm states it is free to download and use, as a UI and CLI for Windows, macOS and Linux hosts.

Which of these matters for embedded and edge teams?

Most embedded projects will touch only one of the three directly. If you are deploying AI on embedded NPUs or edge Linux devices, the Arm AI Portal is the relevant entry point. Mobile game and graphics teams should look at Mali G2-Ultra NX and the Neural Graphics Development Kit, and teams running services on Arm servers should try Performix.

None of them replaces the tools used to bring up and debug hardware. For Cortex-M and Cortex-A firmware work, Pertech supplies and supports Arm’s embedded development tools in Israel:

Building on Arm in Israel? Pertech is an Arm representative in Israel. Tell us about your target device, toolchain and debug needs and we will help you choose the right Arm tools and licenses. Contact Pertech or browse more guides in the Pertech Knowledge Center.

Where to start

  1. Browse models and workflows on Arm AI Portal and read the launch announcement.
  2. Explore the Arm Neural Technology and Neural Graphics Development Kit page, and the Mali G2-Ultra NX technical blog.
  3. Download Arm Performix and profile a workload on an Arm Neoverse instance.

Frequently Asked Questions

What is Arm AI Portal?

Arm AI Portal is a platform from Arm that helps developers and AI agents discover, optimize and deploy AI software on Arm-based compute, from cloud to edge. It offers pre-optimized models with performance data, deployment-ready code and optimization tools, and is available at developer.arm.com/ai. Launch models include Alibaba Qwen, Google Gemma and Ultralytics YOLO, with Arm-optimized models also published on Hugging Face.

Can I use Arm AI Portal from a coding agent?

Arm says Arm AI Portal resources can be brought into coding agents through MCP (Model Context Protocol), and the Arm MCP Server lists Claude and GitHub Copilot among supported agents. Arm describes the agent-ready resources as early access ahead of general release, so check the portal for current status.

Is Arm Performix free, and where does it run?

Arm states that Performix is free to download and use. It analyzes workloads running on Arm Neoverse-based Linux systems, including AWS Graviton, Microsoft Cobalt and Google Axion, and is available as a UI and CLI for Windows, macOS and Linux hosts (Arm64 and x86).

When will devices with the Mali G2-Ultra NX GPU be available?

The material Arm published for the launch does not give a device shipping date. Mali is silicon IP that chip makers integrate into their own SoCs, so availability depends on those partners. For now the Arm Neural Graphics Development Kit is the practical starting point for developers.

How does this relate to Keil MDK, Arm Development Studio and ULINK or DSTREAM-ST?

These announcements cover AI models, mobile graphics and cloud profiling. Bringing up, debugging and tracing Arm-based embedded hardware is a different job, handled by tools such as Keil MDK, Arm Development Studio and the ULINK and DSTREAM-ST debug probes. As an Arm representative in Israel, Pertech supports those tools locally.

Arm, Mali, Neoverse, Keil and Arm Create are trademarks of Arm Limited (or its subsidiaries) in the US and/or elsewhere. Qwen, Gemma, Ultralytics YOLO, Hugging Face and other names are trademarks of their respective owners. Performance figures are Arm’s own claims, taken from the Arm pages linked above; actual results may vary. Information reflects Arm’s announcements dated 8 September 2026.


How to Use Zephyr RTOS with Arm Keil MDK

Zephyr RTOS gives you a modern, permissively-licensed real-time operating system with a huge board and driver ecosystem, built and configured through its own west/CMake toolchain. Arm Keil MDK gives you µVision’s source-level debugger, RTX-class breakpoints and watch windows, and — through ULINK2, ULINKplus, ULINKpro and DSTREAM-ST — instruction trace, code coverage and execution profiling that Zephyr’s own command-line debug tooling does not provide. The two are not mutually exclusive: you can build a Zephyr application entirely with Zephyr’s own build system, then hand the resulting ELF straight to µVision for debugging, without ever creating a Keil build project. This guide walks through exactly how.

What Is Zephyr RTOS?

Zephyr is an open-source, real-time operating system hosted by the Linux Foundation, designed for resource-constrained and embedded devices. It supports a very wide range of Arm Cortex-M, Cortex-A/R and other architectures, and is built with west (Zephyr’s meta-tool) driving a CMake + Ninja/Make build.

What Is Keil MDK?

Arm Keil MDK is Arm’s own microcontroller development kit — the µVision IDE, Arm Compiler, and CMSIS software packs — paired with Arm Keil ULINK and DSTREAM-ST debug probes. MDK is not tied to any single RTOS or build system: µVision can debug any ELF/AXF image that carries standard debug information, regardless of which toolchain produced it.

Why Debug a Zephyr Application with Keil MDK?

Zephyr’s own west debug and west debugserver commands drive GDB through runners such as pyOCD, J-Link or OpenOCD — Arm’s own ULINK and DSTREAM-ST probes are not among the supported runners. If your team already owns Arm debug hardware, or needs capabilities Zephyr’s default debug flow does not offer — instruction trace, code coverage, execution profiling, or Arm Development Studio/µVision’s debug UI — the practical path is to build with Zephyr as usual, then load the resulting image directly into µVision as a “debug-only” session.

Prerequisites

  • A working Zephyr development environment (Zephyr SDK, west, and Python dependencies installed per the official Getting Started Guide)
  • Keil MDK (µVision 5) installed, with the CMSIS device pack for your target microcontroller
  • An Arm Keil ULINK2, ULINKplus, ULINKpro or DSTREAM-ST probe, or a CMSIS-DAP adapter, wired to your board’s JTAG/SWD header

Step 1 — Build Your Zephyr Application

Build normally with west, targeting your board:

west build -b <your_board> path/to/your/app

By default Zephyr’s CMake build compiles with debug information enabled, which is what makes source-level debugging in µVision possible afterwards. The resulting debug image is written to build/zephyr/zephyr.elf (Zephyr also produces zephyr.hex and zephyr.bin, but only the ELF carries the symbol and line-number information µVision needs for source-level debug).

Step 2 (Optional) — Build with the Arm Compiler 6 Toolchain

Zephyr defaults to the GNU Arm Embedded toolchain, but it also supports Arm Compiler 6 (armclang) — the same compiler used by Keil MDK — as a third-party toolchain. To use it, set before building:

export ZEPHYR_TOOLCHAIN_VARIANT=armclang
export ARMCLANG_TOOLCHAIN_PATH=/path/to/your/arm/compiler
export ARMLMD_LICENSE_FILE=/path/to/license_armds.dat

This requires a valid Arm Compiler license (e.g. from an MDK-Professional or Arm Development Studio installation). It is not required to debug in µVision — a GNU-built ELF debugs in µVision perfectly well — but keeping build and debug on the same compiler family avoids any codegen differences between what you build and what you step through.

Step 3 — Create a Debug-Only µVision Project

Rather than trying to recreate Zephyr’s build inside µVision, create a minimal project whose only job is to load and debug the Zephyr-built ELF:

  1. In µVision, Project › New µVision Project and select the exact microcontroller your Zephyr board targets (matching device is what supplies the correct memory map and flash algorithm).
  2. When prompted for startup/runtime files, you can decline adding them — this project will not compile any source, it only hosts debug settings.
  3. Open Project › Options for Target › Debug, select your debug adapter (ULINK2 / ULINKplus / ULINKpro / DSTREAM-ST / CMSIS-DAP) from the dropdown, and configure it under Settings (port SW, correct clock speed for your target).
  4. Still on the Debug tab, tick Load Application at Startup and Run to main(), and untick Update Target before Debugging under the Utilities tab if present, so µVision does not try to (re)build anything before launching the debug session.

Step 4 — Point µVision at the Zephyr ELF

In Options for Target › Debug, browse to your build output as the image to load: build/zephyr/zephyr.elf. µVision’s debugger reads the ELF’s DWARF debug information directly, so source files, line numbers, function names, global and static variables all resolve normally — you can also drive this via the LOAD command from the command window, e.g. LOAD build\zephyr\zephyr.elf, if you prefer to load images ad hoc without changing project settings.

Step 5 — Start Debugging

Click Debug › Start/Stop Debug Session. µVision resets the target, downloads the Zephyr image over your probe, and stops at main() (or wherever your entry breakpoint is set). From here you have full source-level debugging of your Zephyr application: breakpoints, step/step-over/step-into, local and global variable watch, memory and register views, and — if you’re using ULINKpro or DSTREAM-ST — instruction trace, code coverage and execution profiling against the running Zephyr image, the same as any other Cortex-M target.

Tips for a Clean Debug Session

  • Keep optimization at -O0 or -Og in your Zephyr build (via your application’s CMakeLists.txt or a debug build type) for reliable source-line stepping — higher optimization levels reorder and inline code in ways that make step-by-step debugging confusing regardless of which debugger you use.
  • Double-check the device selected in your µVision project matches your Zephyr board’s exact part number, so the flash algorithm and memory map line up.
  • µVision does not ship a dedicated Zephyr thread-awareness view the way it does for Keil RTX — you get full Cortex-M/CoreSight-level debugging (registers, memory, call stack, trace), but not a built-in “list of Zephyr threads” panel. Zephyr’s thread structures can still be inspected manually via the Watch/Memory windows if needed.
  • Re-run west build and reload the ELF (or restart the debug session) after every source change — this workflow does not auto-rebuild from µVision.

Frequently Asked Questions

Can ‘west debug’ drive a Keil ULINK or DSTREAM-ST probe directly?

No. Zephyr’s west debug/west debugserver commands support runners such as pyOCD, J-Link and OpenOCD, not Arm’s ULINK/DSTREAM-ST family. To use those probes, build with west as usual and load the resulting ELF into a µVision debug session instead, as described above.

Do I need an Arm Compiler license to debug a Zephyr build in µVision?

No. µVision can debug an ELF built with the default GNU Arm Embedded toolchain with no extra license — an Arm Compiler 6 (armclang) license is only needed if you choose to build with ZEPHYR_TOOLCHAIN_VARIANT=armclang instead.

Does this give me Zephyr thread-aware debugging in µVision?

Not out of the box. µVision provides full Cortex-M source-level debugging, memory/register access, and — with ULINKpro or DSTREAM-ST — instruction trace and code coverage against the Zephyr image, but it does not include a built-in Zephyr-specific thread-list view the way it does for Keil RTX.

Pertech supplies the full Arm Keil MDK toolchain and the ULINK2, ULINKplus, ULINKpro and DSTREAM-ST debug probes used in this workflow, with local stock and technical support in Israel. Contact us if you need help wiring up Zephyr and Keil MDK for your project.


upbtn
Skip to content