Description
Linux with PREEMPT_RT, now part of the mainline kernel, can deliver low and bounded latencies on suitable hardware, but only when the kernel, firmware, drivers and application are configured with care. Default distributions are tuned for throughput, not determinism. This service takes your Linux target from unknown latency to measured, tuned and documented real-time behavior.
What’s included
- Baseline latency measurement using standard tools such as cyclictest under representative and stress loads.
- Kernel configuration review and build guidance for PREEMPT_RT, including timer, tick and power-management options.
- System tuning: CPU isolation, IRQ affinity, threaded interrupt priorities, frequency scaling and idle-state policy, and memory locking.
- Firmware and BIOS settings review where applicable, such as power-saving states that can add latency.
- Scheduling setup for your critical threads using SCHED_FIFO or SCHED_DEADLINE reservations, with guidance on choosing runtime, deadline and period parameters.
- Identification of drivers or kernel paths causing latency spikes, using kernel tracing tools.
Deliverables
- A latency report comparing baseline and tuned results, with test conditions documented.
- Kernel configuration fragments, boot parameters and system configuration scripts.
- Recommendations for application structure, such as thread priorities, memory allocation and I/O handling.
- A review call to hand over the configuration and answer questions.
Typical timeline
One to two weeks from scope confirmation for a single hardware platform. Additional platforms or custom driver investigation can be added in the statement of work.
Who it’s for
Teams using Linux for robotics, motion control, industrial gateways, audio and video processing, or test and measurement equipment, who need tighter timing than a standard kernel provides. It also helps teams deciding whether Linux can meet their requirements at all, or whether a dedicated RTOS or hybrid architecture is a better fit; see Choosing an RTOS in 2026. The achievable latency depends on your hardware, so we report what we measure rather than promise a figure in advance.
How it works
- Order: place your order on this page. You will receive an order confirmation by e-mail.
- Kick-off call: we contact you within a few business days to schedule a video call to discuss your system, goals and constraints.
- Scope confirmation: we confirm the scope, deliverables, schedule and any confidentiality arrangements in a written statement of work before substantive work begins.
- Delivery: we carry out the work as agreed and walk you through the results.
The price shown covers the standard scope described above. The final scope is always confirmed in the statement of work; if your system needs significantly more effort, we will tell you before any additional cost is incurred. See our refund policy and terms and conditions for details.
