Credit Parameter Flow Control for Packet Admission
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing processor systems require large buffer capacities due to uncontrolled admission of incoming data traffic, leading to potential data bursts and increased downstream buffer requirements.
Innovation Solution
A method that adjusts the first credit parameter based on the value of a second credit parameter, limiting credit value increases and decreases to prevent large credit builds and reduce burst sizes, thereby reducing buffer capacity requirements by adapting data admittance to processing resources such as bit rate and packet rate capacities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data packets are admitted as quickly as possible without controlled admittance limitation, then processing speed is improved, but buffer capacity requirements increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing credit values in token buckets before data admission decisions are made. Each shaper maintains credit values that are periodically increased, and data packets are admitted only when credit values are sufficient. This advance preparation of credit resources enables fast admission decisions without requiring large buffers to handle admission control delays.
2Adaptability or versatility
If multiple shapers are used to control traffic based on different resources (bit rate capacity and packet rate capacity), then resource management is improved, but data bursts occur when one credit value builds up while another is depleted
Solution Approach 1:
The patent implements feedback by continuously monitoring the credit values of multiple token buckets and using this information to control data admission. When credit values are depleted due to packet admission, the system periodically replenishes them based on the respective resource capacities (bit rate and packet rate). This feedback mechanism ensures that admission decisions reflect current resource availability, preventing bursts that would occur if one credit value builds up while another is depleted.
3Stability of the object's composition
If credit values are increased periodically with predetermined amounts, then constant bit rate and packet rate can be achieved, but large buffer capacity is required downstream to handle potential bursts
Solution Approach 1:
The patent applies dynamics by making the data admission process adaptive rather than static. Instead of using fixed admission rates, the system dynamically adjusts admission decisions based on current credit values in the token buckets. The credit values are periodically increased but decrease when packets are admitted, creating a dynamic balance that maintains constant average bit and packet rates while preventing large bursts. This dynamic control eliminates the need for large downstream buffers.
Data Source
AI summary
The present invention relates to a processor and a method for processing a data packet, the method including steps of decreasing a value of a first credit parameter when the data packet is admitted to a processor at least partly based on the value of the first credit parameter and a first limit of the first credit parameter, and increasing the value of the first credit parameter, in dependence on a data storage level in a buffer in which the data packet is stored before being admitted to the processor, the value of the first credit parameter not being increased, so as to become larger than a second limit of the first credit parameter, when the buffer is empty.