Object Memory Fabric Distributes Data Across Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current commodity hardware and software solutions are inadequate in managing processing, memory, and storage efficiently, leading to inefficiencies and limitations in Cloud Computing and Big Data environments due to the separation of these components and the complexity of software stacks, which results in suboptimal performance and economic scaling.
Innovation Solution
Implementing an object-based memory fabric that distributes and tracks memory objects across a hierarchy of processing nodes, allowing each node to manage memory objects natively without I/O instructions, and using a unified algorithm for object management to create a larger cache and reduce bandwidth requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If processing, memory, and storage are managed separately by software on each node, then each component can be independently managed, but the system complexity increases and performance efficiency decreases
Solution Approach 1:
The patent merges processing, memory, and storage management into a unified hardware-based object memory fabric where these components operate as an integrated system rather than separate software-managed entities. This eliminates the complex software stack while maintaining independent manageability through hardware-level operations.
Solution Approach 2:
The patent introduces object routers as hardware intermediaries that manage communication and coordination between processing nodes, memory modules, and storage components within the object memory fabric, replacing complex software management with dedicated hardware mediation.
2Adaptability or versatility
If data is accessed through multiple layers of software stack, then data can be managed with complex protocols, but access latency increases and efficiency decreases
Solution Approach 1:
The patent extracts data management functionality from the complex software stack and implements it directly in hardware within the object memory fabric. This removes multiple software layers that cause latency while preserving adaptability through hardware-configurable object routing and management capabilities.
3Reliability
If memory objects are distributed across multiple nodes, then system scalability and fault tolerance improve, but bandwidth requirements and management complexity increase
Solution Approach 1:
The patent implements dynamic object routing within the hardware-based object memory fabric that intelligently distributes memory objects across nodes based on current system state, access patterns, and availability. This dynamic distribution achieves fault tolerance and scalability while optimizing bandwidth usage through hardware-level coordination.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
According to one embodiment, a hardware-based processing node of a plurality of hardware-based processing nodes in an object memory fabric can comprise a memory module storing and managing a plurality of memory objects in a hierarchy of the object memory fabric. Each memory object can be created natively within the memory module, accessed using a single memory reference instruction without Input/Output (I/O) instructions, and managed by the memory module at a single memory layer. The object memory fabric can distribute and track the memory objects across the hierarchy of the object memory fabric and the plurality of hardware-based processing nodes on a per-object basis. Distributing the memory objects across the hierarchy of the object memory fabric and the plurality of hardware-based processing nodes can comprise storing, on a per-object basis, each memory object on two or more nodes of the plurality of hardware-based processing nodes of the object memory fabric.