Remote Server Brokerage Based on Performance Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote access technologies provide limited information for selecting servers, leading to poor server utilization and user difficulty in choosing suitable servers for demanding applications, as users lack knowledge about server performance and installed applications.
Innovation Solution
Implementing a performance-based brokering system where a broker collects and aggregates performance data from servers, ranks them based on application-specific metrics, and directs clients to the best-performing servers for remote desktop connections, providing informed server selection and improved utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If static information (operating system, hostname, power state) is provided for server selection, then the information is easy to obtain, but users cannot accurately determine which server will be suitable for their specific application
Solution Approach 1:
The broker performs preliminary actions by collecting, storing, and pre-processing performance data from multiple servers before users need to make selection decisions. This includes aggregating historical performance metrics and application compatibility information so that when a user selects an application, the broker can immediately provide relevant server recommendations without requiring real-time complex analysis.
Solution Approach 2:
The broker acts as an intermediary between users and servers, translating user application selections into optimized server recommendations. It mediates the information flow by collecting performance data from servers, processing it according to application requirements, and presenting simplified, actionable recommendations to users, thereby reducing the information loss without exposing users to system complexity.
2Ease of operation
If users manually select servers based on limited information, then the selection process is simple, but server utilization is poor and users experience difficulty in choosing suitable servers
Solution Approach 1:
The system implements feedback mechanisms where performance data from server operations is continuously collected and fed back into the broker's database. This feedback loop enables the broker to learn from actual server performance patterns and improve its recommendations over time, thereby enhancing server utilization efficiency while keeping the user interface simple and ease of operation intact.
Solution Approach 2:
The broker provides self-service functionality by automatically processing server performance data and generating personalized server recommendations based on user application selections. This eliminates the need for users to manually analyze complex server information, maintaining ease of operation while significantly improving server utilization efficiency through automated, intelligent matching.
3Measurement precision
If performance data is collected and processed for each application, then accurate server recommendations are achieved, but the processing time and system resources increase
Solution Approach 1:
The system performs preliminary data processing by continuously collecting and pre-processing performance metrics from servers in advance. Historical performance data is aggregated, cleaned, and organized in the broker's database before users need recommendations, so that when an application is selected, the system can quickly retrieve and present relevant information without time-consuming real-time analysis.
Solution Approach 2:
The broker employs partial action by selectively processing only the performance data relevant to the specific application being accessed. Rather than processing all possible server metrics for all applications, the system identifies and processes only the critical performance parameters needed for that particular application, thereby maintaining measurement precision while reducing unnecessary processing time and resource consumption.
Data Source
AI summary
Examples of a method for brokering remote servers are described herein. In some examples, performance data is received from a plurality of remote servers, where the performance data indicates rendering performance of a foreground application executed by at least one of the remote servers and streamed from at least one of the remote servers over a remote desktop connection. An indication of a selected application is received from a client. The client is directed to at least one of the remote servers based on the performance data and the selected application.