Edge-node Automated Media Processing

OVERVIEW & GOAL

Several inxware users were looking to migrate their edge-IoT compute hardware platform to the Arm Cortex-A-series silicon. As their service businesses thrived, many began to look at cost optimizations and other scaling factor opportunities that Arm-based systems can offer, such as:
       — Reduced unit costs
       — Improved component life-cycle, availability, and consistency
       — Improved reliability
       — Reduced energy consumption and thermal management
       — Improved physical form-factor options

A particularly active sector of users requires a common set of features that they were looking to optimize to extend the scope of their offering:
       — Media data compute and real-time IO
       — Real-time generative AI
       — Content distribution and scheduling
       — IoT connectivity
       — Multi-OS requirements (including Linux & Android)

The hardware solution roadmap had to make the best use of modular Arm compute products and provide customization for application-specific IO and product form-factor considerations.

CHALLENGE

Priority requirements for clients to engage with the proposed transition to new hardware and sometimes new operating systems included:
       — Avoidance of repeating previous software NRE
       — Maximum exploitation of legacy field-proven application software
       — Seamless compatibility of legacy and new edge devices on their platforms with minimal (if any)                  need to differentiate

Many of the client’s inxware-lucid applications involve 1,000s of no-code function blocks (equating to 100,000s of lines of conventional code). Much of this application complexity deals with real-world scenarios and exception handling that is required to provide high-quality 24/7/365 services to customers.

The value of carrying the software capability maturity models (CMM) achievements through to new products is extremely important to service providers. Retaining the trust of customers and minimizing the support costs of rolling out new products is business critical to many IoT-enabled services.

SOLUTION & BENEFITS

inxware enabled users to continuously build on their innovations by exploiting the quad-core power of Arm Cortex-A53 silicon and the availability of Arm-optimized OSs. The Lucid environment helped ensure
their existing high-quality, self-documenting software worked efficiently from the outset and avoided developers encountering technical debt challenges during the transition.

Product developers were easily able to maintain a single application across legacy and new platforms when it came to rolling out and supporting their new products. The single application approach worked
across many different physical product specifications, which were effectively mitigated using inxware’s stub function approach to handle unmet dependencies appropriately at runtime. inxware users benefited throughout their product evolution where they were able to implement radical changes to their products by transitioning to Arm-based IP with minimal risks and costs, while also addressing new market opportunities

Read more about our ARM-based projects here.