Web Server Image Transmission via Control Channel
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for transmitting image data in computing systems, such as VDI, face latency overhead due to multiple HTTP GET requests and incur increased bandwidth consumption when using text-based encoding, with limited support for deflate compression in websockets implementations.
Innovation Solution
A system and method where a web server suppresses responses to image requests until generated images are received, transmitting them via a control channel based on triggering events, reducing latency and bandwidth impact by using a control channel in conjunction with web request channels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple HTTP GET requests are used to retrieve image data, then the web client can receive images from the web server, but latency overhead increases due to multiple sequential requests
Solution Approach 1:
The patent implements a continuous image data stream transmission mechanism where the web server sends images sequentially without requiring the client to initiate each request. The server pushes images to the client using a keep-alive connection with chunked encoding, eliminating the stop-start nature of multiple HTTP GET requests and reducing latency overhead.
Solution Approach 2:
The web server establishes a keep-alive connection and prepares to push image data proactively before the client needs to request each image. By pre-establishing the connection and using server-side pushing mechanisms, the system eliminates the waiting time associated with sequential client-initiated requests.
2Loss of time
If text base64 encoding is used to push image data from web server to web client, then latency overhead is reduced by eliminating client requests, but bandwidth consumption increases
Solution Approach 1:
The patent changes the encoding parameter from text-based base64 to binary format for image data transmission. This parameter change reduces the data size by approximately 33% compared to base64 encoding, thereby reducing bandwidth consumption while maintaining the low-latency push mechanism.
Solution Approach 2:
The patent uses a universal binary image format that can be directly transmitted over the keep-alive connection without requiring conversion to text format. This multi-functional approach allows the same transmission mechanism to handle both the push protocol and the image data efficiently in binary form.
3Quantity of substance
If deflate compression is applied to reduce bandwidth consumption, then bandwidth is optimized, but additional latency is introduced for encoding and decoding
Solution Approach 1:
The patent extracts the compression step from the transmission pipeline by sending image data in an already-compressed binary format from the server. This eliminates the need for real-time compression and decompression operations during transmission, thereby avoiding the latency overhead associated with these operations while still achieving bandwidth optimization.
4Quantity of substance
If websockets are used to deliver binary image data, then bandwidth consumption is reduced, but most implementations do not support deflate compression and conversion to base64 is required
Solution Approach 1:
The patent uses a simplified copy mechanism where binary image data is directly pushed from the server to the client through the keep-alive connection without requiring format conversion. This copying approach maintains the efficiency of binary transmission while avoiding the complexity and latency of base64 conversion that would be required in websocket implementations.
Data Source
AI summary
A plurality of web requests for images are received from a web client via corresponding web request channels. A response to a first one of the web requests is suppressed until a first generated image is received by the web server. The first web request is associated to the first generated image, which is then transmitted to the web client as a response to the first web request via the corresponding web request channels. A second received generated image is transmitted to the web client via a control channel based on a detection of a triggering event.