Hybrid Queue Structure for Unlimited Depth Data Transfer
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing multi-tasking systems face challenges in managing bursts of messages from multiple processors, leading to processor stalls due to shallow hardware queues, and existing solutions that overflow into system memory occupy valuable space and do not fully eliminate stalling.
Innovation Solution
A hybrid queue structure that dynamically allocates memory for messages as nodes of a linked list, allowing the queue to overflow into system memory only when full, using a hardware queue with an auxiliary register to manage overflow and flags for synchronization, enabling efficient message processing without reserving unused memory space.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the hardware queue depth is increased to avoid processor stalls, then the stalling limit is pushed back, but the queue occupies an unreasonable surface area on the circuit
Solution Approach 1:
The queue is segmented into a shallow hardware queue portion and a software-managed overflow portion in system memory. The hardware queue contains only enough slots to trigger overflow conditions, while the linked list structure in system memory provides the additional depth needed to avoid processor stalls without expanding hardware queue area.
Solution Approach 2:
The queue depth is extended from the hardware domain to the software/memory domain. By using a linked list structure in system memory that can be dynamically allocated, the queue effectively gains unlimited depth without requiring proportional hardware resources, transitioning the solution from spatial expansion to hierarchical organization.
2Productivity
If the hardware queue depth is increased to avoid processor stalls, then data production continuity is improved, but the hardware complexity and resource consumption increase
Solution Approach 1:
A hybrid queue management mechanism acts as an intermediary between the hardware queue and system memory. This mechanism includes overflow detection logic, linked list construction, and pointer management that coordinates between hardware and software components, enabling continuous data production without requiring the hardware queue to handle all overflow conditions alone.
3Reliability
If a reserved memory area is used for queue overflow, then the stalling limit is pushed back, but unused space is permanently removed from available system memory
Solution Approach 1:
The queue overflow storage transitions from a static reserved memory area to a dynamic linked list structure. Memory pages are allocated on-demand as the queue fills and overflow occurs, and freed when queue slots become available. This dynamic allocation allows the same memory to serve both queue overflow and general system purposes at different times, maximizing memory utilization while maintaining reliability.
Data Source
AI summary
A method for transferring messages from a producer element to a consumer element uses a memory shared between the producer element and the consumer element, and a hardware queue including several registers designed to contain addresses of the shared memory. The method includes the steps of storing each message for the consumer element in the shared memory in the form of a node of a linked list, including a pointer to a next node in the list, the pointer being initially void, writing successively the address of each node in a free slot of the queue, whereby the node identified by each slot of the queue is the first node of a linked list assigned to the slot, and when the queue is full, writing the address of the current node in memory, in the pointer of the last node of the linked list assigned to the last slot of the queue, whereby the current node is placed at the end of the linked list assigned to the last slot of the queue.