Distributed Shared Memory Cache Manager for Data Redundancy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current information handling systems face challenges in efficiently processing and storing data across multiple devices, particularly in managing shared memory resources in a distributed network, which affects data access speed and reliability.
Innovation Solution
A distributed shared memory system is implemented, where physical memory from multiple computing systems is aggregated and managed by local cache managers, allowing clients to access a unified logical memory resource through a network, with features like RDMA, metadata services, and policy managers to optimize data placement and access.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If data is cached in a distributed shared memory system across multiple computing systems, then data access speed is improved, but system complexity increases
Solution Approach 1:
The patent divides the distributed shared memory system into multiple computing systems, each with its own local cache manager. Each cache manager independently manages a portion of the shared memory, allowing parallel data access across multiple nodes while distributing system complexity across separate manageable units rather than concentrating it in a single centralized controller.
Solution Approach 2:
The patent introduces cache managers as intermediary components between clients and the physical memory resources. These cache managers handle memory allocation, caching, and eviction operations, shielding clients from the underlying system complexity while enabling fast data access through intelligent caching strategies.
2Productivity
If physical memory from multiple computing systems is aggregated into a unified logical memory resource, then memory utilization efficiency is improved, but data access reliability challenges arise
Solution Approach 1:
The patent implements a redundancy mechanism where data is copied across multiple computing systems in the distributed shared memory. When data is written to the shared memory, it is replicated to multiple nodes, ensuring that if one node fails or data becomes corrupted, redundant copies remain available, thus maintaining data access reliability while utilizing aggregated memory resources.
Solution Approach 2:
The patent employs error checking and data validation mechanisms that are built into the memory management process before failures occur. By implementing redundancy and validation in advance, the system cushions against potential reliability issues that arise from aggregating memory across multiple systems.
3Productivity
If cache managers are implemented to manage data placement and eviction, then caching efficiency is improved, but operational complexity increases
Solution Approach 1:
The patent implements cache managers that autonomously make decisions about data placement, caching, and eviction based on predefined policies and algorithms. The cache managers self-manage their local memory resources without requiring manual intervention, automatically determining which data to cache, where to place it, and when to evict it, thereby improving caching efficiency while keeping operational complexity manageable through automation.
Solution Approach 2:
The patent allows cache management parameters such as cache size, eviction policies, and data placement strategies to be dynamically adjusted based on system conditions and workload characteristics. This flexibility enables the system to optimize caching efficiency for different scenarios without requiring complete redesign or manual reconfiguration of the entire system.
Data Source
AI summary
In one embodiment, a computing system includes a cache and a cache manager. The cache manager is able to receive data, write the data to a first portion of the cache, write the data to a second portion of the cache, and delete the data from the second portion of the cache when the data in the first portion of the cache is flushed.