Blade PC Host Status Detection via Service Manager
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Managing complex resource allocation in large-scale networked computer systems, particularly in centralized blade PC systems, is challenging due to the inefficiency of existing methods for detecting the status of hosts and I/O peripherals, such as broadcast packages being blocked by routers and time-consuming IP address scanning, which leads to delayed data transmission and server burden.
Innovation Solution
A centralized computer system design that includes a service manager, DHCP server, and DNS server, where hosts and I/O peripherals send request signals with domain names, allowing for automatic status detection and periodic notification of current status, with dynamic IP address updates through DDNS, and utilizing a Modular Management Board to consolidate status updates and reduce network load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If broadcast packages are used for status detection, then all devices in the network segment can receive the status request, but the packages are blocked by routers and cannot cross network segments
Solution Approach 1:
The patent introduces a service manager as an intermediary component that hosts a dynamic domain name service. This service manager acts as a mediator between hosts and the status detection mechanism, allowing status requests to be routed through domain names rather than direct broadcast packages. The service manager receives status requests from hosts, forwards them to the appropriate I/O peripherals, and returns the status information, thereby enabling cross-network-segment communication without relying on broadcast packages that are blocked by routers.
2Reliability
If IP address scanning is performed regularly for status detection, then all devices can be searched, but it consumes more time and causes heavy burden on the server
Solution Approach 1:
The patent implements a self-service mechanism where hosts automatically send status requests to the service manager using domain names, eliminating the need for the server to perform time-consuming IP address scanning. Each host maintains awareness of the service manager's domain name and proactively queries status information, shifting the detection burden from centralized server scanning to distributed host-initiated requests. This approach significantly reduces detection time and server load while maintaining comprehensive device status monitoring.
3Reliability
If IP address scanning is performed regularly, then device status can be detected, but data transmission is delayed and the server suffers heavy burden
Solution Approach 1:
The patent employs preliminary action by pre-configuring hosts with the service manager's domain name through DHCP or manual configuration. This allows hosts to immediately send status requests using domain names without needing to perform time-consuming IP address resolution or scanning at the time of status detection. The domain name system is set up in advance, enabling fast and direct communication between hosts and the service manager, thereby improving data transmission speed while maintaining accurate status detection.
Data Source
AI summary
The invention discloses a centralized computer system, i.e., so-called blade PC system. The centralized computer system according to the invention includes a plurality of centralized and bladed hosts and a plurality of I/O peripherals. In particularly, the centralized computer system according to the invention is capable of automatically detecting a current status of each host and/or each set of I/O peripherals.