Proximity-Based Memory Allocation for Un可靠 Packet Transport
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In distributed shared memory systems with unreliable packet transport, existing memory allocation schemes fail to efficiently scale to large systems and handle non-uniform memory access times, leading to performance issues due to reliance on reliable packet transport assumptions.
Innovation Solution
A proximity-based memory allocation scheme where each processor node broadcasts memory requests to nearby memory nodes, increasing the search radius and timeout values until a suitable memory node is found, using a four-way handshake protocol to maintain interaction and handle unreliable transport layers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a centralized memory allocation scheme is used, then memory allocation for all processors can be managed, but the system cannot efficiently scale to large systems
Solution Approach 1:
The patent divides the centralized allocation function into distributed components. Each processor node independently performs memory allocation by broadcasting requests to nearby memory nodes and selecting from responses, eliminating the need for a single centralized allocator and enabling system scalability.
Solution Approach 2:
The patent introduces a spatial dimension to memory allocation by organizing memory nodes in proximity groups and having processors allocate from nearby memory first. This proximity-based approach adds a spatial organization layer that enables efficient scaling without centralized coordination.
2Reliability
If reliable packet transport is assumed, then memory allocation can proceed, but the system fails in unreliable transport conditions
Solution Approach 1:
The patent implements preliminary actions to handle potential packet loss before it affects allocation. Timeout mechanisms are established in advance, and processors are prepared to retransmit requests or select alternative memory nodes if responses are not received, ensuring allocation proceeds reliably even in unreliable networks.
Solution Approach 2:
The patent prepares compensatory measures in advance for unreliable transport by implementing timeout thresholds and alternative selection criteria. If initial requests are lost or delayed, the system has pre-established fallback mechanisms to continue allocation without failure.
3Quantity of substance
If distant memory nodes are selected, then more memory capacity is available, but access latency increases
Solution Approach 1:
The patent applies local quality by organizing memory allocation in proximity-based groups. Processors first attempt to allocate from nearby memory nodes, and only expand to distant nodes when local capacity is exhausted. This creates a quality gradient where local allocations are faster and distant allocations provide additional capacity.
Solution Approach 2:
The patent implements dynamic allocation where the search radius and timeout values are adjusted based on system conditions. The system dynamically expands from local to distant memory nodes as capacity is needed, balancing latency and available capacity adaptively rather than using a fixed allocation strategy.
Data Source
AI summary
A system and method for allocating physical memory in a distributed, shared memory system and for maintaining interaction with the memory using a reservation protocol is disclosed. In various embodiments, a processor node may broadcast a memory request message to a first subset of nodes connected to it via a communication network. If none of these nodes is able to satisfy the request, the processor node may broadcast the request message to additional subsets of nodes until a positive response is received. The reservation protocol may include a four-way handshake between the requesting processor node and a memory node that can fulfill the request. The method may include creation of a reservation structure on the requesting processor and on one or more responding memory nodes. The reservation protocol may facilitate the use of a proximity-based search methodology for memory allocation in a system having an unreliable underlying transport layer.