Unified Memory Hierarchy for Mobile Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computing devices, such as mobile devices, face limitations in cache and system memory due to size, cost, and thermal constraints, leading to insufficient memory resources for running multiple applications and files simultaneously, resulting in potential shutdowns and user dissatisfaction.
Innovation Solution
The user device extends its memory resources by combining cache memory, system memory, and storage resources, including flash memory and network resources, to create a unified addressable information space, allowing it to access and store data across various levels of memory hierarchy, including storage devices and network servers, to accommodate memory-intensive applications and files.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If the size of cache and system memory is increased to accommodate multiple applications and files, then memory resources are improved, but device size, cost, and thermal constraints are worsened
Solution Approach 1:
The patent extends the memory hierarchy by adding a new dimension - network storage resources accessible via network interfaces. This creates a multi-layered memory architecture that spans from local cache memory through system memory to remote network storage, effectively increasing available memory resources without physically expanding the device footprint.
Solution Approach 2:
The patent implements a nested memory hierarchy where cache memory is nested within system memory, which is nested within storage memory, which is nested within network storage. Each layer provides additional capacity and is accessed through the previous layer, creating a consolidated memory space that appears as a single unified resource to applications.
2Adaptability or versatility
If cache and system memory size is increased to run multiple applications simultaneously, then application concurrency is improved, but device size, cost, and thermal constraints are worsened
Solution Approach 1:
The patent makes storage memory and network resources serve dual functions - acting as both long-term storage and as extended cache/memory resources for running applications. The memory manager dynamically allocates these resources based on application needs, allowing the same physical resources to function as either storage or working memory depending on the situation.
Solution Approach 2:
The patent implements dynamic memory management where the allocation of cache, storage, and network resources to applications is not fixed but continuously adjusted based on current system state, application requirements, and performance metrics. This allows the system to adapt memory resources in real-time to support varying application concurrency demands.
3Quantity of substance
If memory resources are extended to storage and network levels, then available memory capacity is improved, but data access time is worsened
Solution Approach 1:
The memory manager proactively pre-loads data from storage and network resources into faster cache and system memory before it is actually needed by applications. By anticipating future access patterns and preparing data in advance, the system reduces actual access time when applications request the data.
Solution Approach 2:
The patent introduces cache memory and system memory as intermediary layers between applications and slow storage/network resources. These intermediaries hold frequently accessed data closer to the processor, mediating between the speed requirements of applications and the capacity/slow speed of extended storage resources.
4Quantity of substance
If storage memory is used as extended cache, then memory resources are improved, but memory hierarchy complexity is worsened
Solution Approach 1:
The memory manager autonomously manages the complexity of the extended memory hierarchy without requiring application awareness or intervention. It automatically handles data placement, eviction, and allocation across cache, storage, and network resources based on usage patterns and system state, shielding applications from the underlying complexity.
Solution Approach 2:
The patent segments the memory management functionality into distinct components - cache management, storage management, network resource management, and a coordinating memory manager. Each segment handles specific aspects of the extended memory hierarchy, making the overall complex system manageable through modular organization of management responsibilities.
Data Source
AI summary
Systems and methods for extending the memory resources of a user device to storage resources and/or network resources associated with the user device. The cache and system memory of the user device may be utilized as a cache memory and the storage resources and/or network resources of the user device may be utilized as a storage memory.