Sinea technology charging | How does the Sinsegye iComputer perform core separation operations?

Hello everyone,

I'm Sinea,

the intelligent customer service assistant of Sinsegye.

Recently, I've received quite a few "little troubles" from engineer partners.

iComputers deployed at key nodes of the production line have good configurations, but why do they occasionally feel less "smooth" in response? Why does the real-time performance of monitoring data fluctuate sometimes? Or why are critical control tasks occasionally "overshadowed" by other background services?

Don't worry,

in many cases, it's not that the hardware is insufficient,

but that the division of labor among CPU cores is not clear enough,

and resource allocation can be more optimized!

This week, I'll bring you a very "hardcore" operation tip!

How to perform CPU core partitioning for Sinsegye iComputers via DeviceManager?

In industrial sites, what we pursue is not only "operability" but also determinism, low latency, and high reliability. The core significance of CPU core partitioning lies in the following:

• Eliminate "resource grab" battles: By default, the operating system dynamically schedules all tasks to run on all cores. This seems reasonable and efficient, but in industrial scenarios, it means that a background log task or a temporary file scan may seize the execution time slices of critical tasks such as motion control and real-time data collection, leading to delay jitter or even timeouts. CPU core partitioning can "pin" critical tasks to dedicated cores, physically isolating interference from other tasks and ensuring they get 100% computing power.
• Reduce switching overhead and speed up response: Task switching between different cores brings additional overhead such as cache invalidation and context switching. Binding specific tasks to fixed cores greatly reduces these overheads, making the execution path of critical tasks shorter and more "responsive," thus improving real-time response capability.
• Ensure the determinism of core businesses: For core industrial applications such as PLC Runtime, real-time control loops, and precise clock synchronization, even the smallest delay is intolerable. CPU core partitioning builds an "isolation zone" for them, ensuring that their execution time is predictable and guaranteed, which is a hard requirement for many high-precision manufacturing scenarios.
• Optimize resource utilization efficiency: Through reasonable isolation and binding, background tasks that are less sensitive to latency (such as file transmission and remote monitoring agents) can be arranged to run on remaining cores. This avoids interfering with core businesses while making full use of hardware resources, achieving the golden balance of overall performance.

Simply put,

CPU core partitioning is like installing "traffic police" and "VIP dedicated lanes" for your iComputer,

keeping core businesses running smoothly!

In edge computing deployments, especially scenarios where real-time control and non-critical services are mixed, this is an essential skill to improve the overall stability and performance of the system!

Sinsegye's iComputer products can perform CPU core partitioning through two methods: local installation of DeviceManager or using HyperCare.For the specific usage of HyperCare, you can visit the following address: SF8010 Operation Help Manual.

This time, we'll explain how to perform CPU core partitioning using the locally installed DeviceManager.

First, you need to visit the official Sinsegye website to download the latest installation package of DeviceManager. You can specifically visit the following address: Sinsegye Resource Download Center.

After installation, follow these steps:

1. Log in to the iComputer 192.168.1.200 via SSH

ote: For some models, in rare cases, you may need to install adi and aoi in advance. Please refer to: DeviceManager Running Component Configuration

2. Open the DeviceManager plug-in manager

3. Click "Install ACP Service" at the bottom left

4. Enter the iComputer's IP, username, and password, then click "Online Installation"

5. Select the installed iComputer

6. Click "Install RTE Plug-in Service"

7. Enter the iComputer's username and password, then click "Online Installation"

8. Installation completed

9. Click "CPU Core Partitioning Management" to view the CPU allocation between non-real-time and real-time domains

Among them, gray indicates CPUs that cannot be reallocated, and blue indicates CPUs that can be allocated.

10. You can drag any blue CPU to the non-real-time or real-time domain. For example, in the figure below, drag blue CPU 2 from the non-real-time domain to the real-time domain.

11. Click "Check Allocation" — select "Confirm" to make it effective after restart.

12. Wait for the restart to complete.