Adaptive Queue Compression for Latency-Sensitive Network Traffic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for processing data packets through intermediary devices often introduce latency and jitter due to improper application of optimization techniques like compression, especially when multiple devices are involved in the network, leading to variations in packet delay over time.
Innovation Solution
The system dynamically selects a subset of packets for processing using optimization techniques, monitoring the queue size and data rates to determine when to apply compression, reserving a portion of the queue for compressed data and allowing other packets to pass without compression, thereby reducing latency and jitter.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If compression is applied to all packets in the queue, then data transmission efficiency is improved, but packet delay variation (jitter) increases
Solution Approach 1:
The queue is divided into two separate queues: a first queue for uncompressed packets and a second queue for compressed packets. This segmentation allows the system to apply compression selectively to certain packets while leaving others uncompressed, thereby improving transmission efficiency for suitable packets while avoiding the introduction of jitter that would occur if all packets were compressed uniformly.
Solution Approach 2:
Instead of applying compression to all packets (excessive action), the system applies compression only to a subset of packets that benefit from it (partial action). The packet selection module determines which packets should be compressed based on specific criteria, applying compression only where it provides net benefit without causing harmful delays to latency-sensitive traffic.
2Productivity
If packets are accumulated in the queue for compression processing, then compression efficiency is improved, but latency is introduced
Solution Approach 1:
The system dynamically adjusts the compression processing approach based on real-time queue conditions and packet characteristics. The packet selection module continuously evaluates which packets to compress and which to send immediately, creating a dynamic balance between achieving compression efficiency and minimizing latency introduction.
Solution Approach 2:
The system performs preliminary evaluation of packets to determine their suitability for compression before adding them to the compressed queue. This preliminary action allows the system to identify packets that can be efficiently compressed without causing excessive delays, making informed decisions about compression processing in advance.
3Productivity
If multiple intermediary devices are deployed, then network optimization is improved, but jitter and latency are amplified
Solution Approach 1:
Each intermediary device segments its traffic into uncompressed and compressed queues, allowing selective optimization at each hop. This segmentation prevents the amplification of jitter that would occur if all devices compressed all packets uniformly, as each device can make local decisions about which packets to compress based on current network conditions.
Solution Approach 2:
The packet selection module acts as an intermediary decision-making component that determines whether each packet should be compressed or sent immediately. This intermediary function allows the system to optimize network traffic flow while preventing the accumulation of delays that would occur with uniform compression at multiple intermediary devices.
Data Source
AI summary
This disclosure is directed to embodiments of systems and methods for performing compression of data in a queue. A device intermediary between a client and a server may determine that a length of time to move existing data maintained in a queue from the queue exceeds a predefined threshold. The device may identify, responsive to the determination, a first quantity of the existing data to undergo compression, and a second quantity of the existing data according to a compression ratio of the compression. The device may reserve, according to the second quantity, a first portion of the queue that maintained the first quantity of the existing data, to place compressed data obtained from applying the compression on the first quantity of the existing. The device may place incoming data into the queue beyond the reserved first portion of the queue.


