# US 2022/0027200 A1 · System for operating and method for arranging nodes thereof

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2022/0027200 A1
- Document ID: US20220027200A1
- Status: Published application
- Kind: A1
- Application number: 17/109,657
- Filed: 2020-12-02
- Published: 2022-01-27
- Earliest claimed: 2020-12-02
- Assignees: National Applied Research Laboratories
- Applicants: National Applied Research Laboratories
- Inventors: CHIN-CHEN CHU; HUNG-FU LU; JHENG YU CHEN; SAN-LIANG CHU; AUGUST CHAO
- Web page: https://patentgrep.com/patent/US20220027200A1

## CPC

- G06F 9/5027: {the resource being a machine, e.g. CPUs, Servers, Terminals}
- G06F 9/5066: {Algorithms for mapping a plurality of inter-dependent sub-tasks onto a plurality of physical CPUs (mappping at compile time, see G06F8/451)}
- G06F 9/4406: {Loading of operating system}
- G06F 9/5038: {considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration (scheduling strategies G06F9/4881 and subgroups)}
- H04L 67/10: in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46)
- G06F 9/455: Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F 2209/5021: Priority
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

The present application reveals a system for computing and a method for arranging nodes thereof, which is applied for a remote host connected with a plurality of computing nodes divided to a plurality of first nodes and second nodes due to a first computing mode and a second computing mode. After the remote host receives a job, the remote host evaluates the computing nodes in accordance with the job and a corresponding priority weight parameter to generate a job beginning data to set the first nodes or the second nodes and to proceed the job. While setting the first or the second nodes, the remote host provides the corresponding system image to the corresponding nodes; while the first or the second nodes are full in resource arrangement, the empty nodes will be converted to the supplement nodes with the corresponding system image from the remote host.

## Claims

### Claim 1

A method for arranging nodes in a system for operating, applied to a remote host of the system connected to a plurality of operating hosts of the system; including the steps of:
inputting an operating task to the remote host by a user end device;
performing an operating resource assessment on the operating hosts in accordance with the operating task and a corresponded priority weight by the remote host, and generating a task initial data;
the remote host judging the operating modes and sequences of the operating hosts according to the task initial data; and
when the task initial data corresponds to a parallel operating mode and performs sorting, the remote host providing a first image data to part of the operating hosts and set them as a plurality of first hosts; when the task initial data corresponds to a logic operating mode and performs sorting, the remote host providing a second image data to part of the said operating hosts and set them into a plurality of second hosts;
wherein when one first resource distribution data corresponding to the first hosts is full, the remote host switches part of the said second hosts into a plurality of first supplementary hosts and provides the first system image data to the first supplementary hosts, letting the first system image data be loaded in the first supplementary hosts; when a second resource distribution data corresponding to the second hosts is full, the remote host switches the said first hosts into a plurality of second supplementary hosts and provides a second system image data to the second supplementary hosts, letting the second system image data be loaded in the second supplementary hosts.

### Claim 2 (depends on 1)

The method for arranging nodes in the system for operating of claim 1, in which the operating hosts are equipped in a distributed computing architecture.

### Claim 3 (depends on 1)

The method for arranging nodes in the system for operating of claim 1, in which the steps that the remote host follows the task initial data to set the first hosts or second hosts include:
When the task initial data corresponds to the secondary operating mode, the remote host further sends the task initial data to one master node in second hosts;
The master node sends the task initial data to one slave node in the second hosts; and
The remote host provides the second system image data to the slave node through the master node, letting the slave node load in the second system image data.

### Claim 4 (depends on 1)

The method for arranging nodes in the system for operating of claim 1, in which the steps that the remote host follows the operating task to assess the operating resource toward the operating hosts along with a corresponded priority weight, the remote host operates the corresponded priority weight according to a queue weight, an operating resource testing parameter, an operating mode and a maintaining parameter of the corresponded operating task.

### Claim 5

A system for operating, which includes:
a plurality of operating hosts;
a remote host, connected to the operating hosts and a database, the database stores a first system image data and a second system image data; the remote host sets the operating hosts as a plurality of first hosts and second hosts according to a parallel operating mode and a logic operating mode; and
an user end device, connected to the remote host, sends an operating task to the remote host; the remote host accesses the operating resource of the operating hosts according to the operating task and a corresponded priority weight and generates a task initial data; the remote host judges the operating modes and sequences of these operating hosts according to the task initial data to; when the task initial data corresponds to the parallel operating mode and performs sorting, the remote host provides a first image data to the first hosts; when the task initial data corresponds to the logic operating mode and performs sorting, the remote host provides a second image data to the second hosts;
when one first resource distribution data corresponding to the first hosts is full, the remote host switches part of the second hosts to a plurality of first supplementary hosts, and sends the first system image data to the first supplementary hosts; when one second resource distribution data corresponding to the second hosts is full, the remote host switches the said first hosts to a plurality of second supplementary hosts and sends the second system image data to the second supplementary hosts.

### Claim 6 (depends on 5)

The system for operating of claim 5, in which the operating hosts are electronic devices with operating capability, each has the heterogeneous hardware with accelerate computing resource, and set in a distributed computing architecture.

### Claim 7 (depends on 5)

The operating system of claim 5, wherein the second hosts include at least one master node and a plurality of slave nodes; the remote host sends the task initial data and the second system image data to the slave node through the master node and load in.

### Claim 8 (depends on 5)

The system for operating of claim 5, wherein the remote host operates the corresponded priority weight according to follows a queue weight, an operating resource testing parameter, an operating mode and a maintaining parameter of the corresponded operating task.

### Claim 9 (depends on 5)

The system for operating of claim 5, wherein the operating host includes:
a role module, equipped with a first role unit and a second role unit; the first role unit corresponds to the first role mode; the second role unit corresponds to the second role mode; the first role unit reads the first image data and loads it into the second role unit;
a first interface, connected to the first role unit;
a second interface, connected to the second role unit; and
a remote interface, connected to the first interface and second interface in serial connections, and connected to the remote host.
