SSD In-Node Processing Unit for Data Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current cloud data storage systems are inefficient due to the need for data to be read from storage nodes, processed by a server CPU, and then stored back, which is costly in terms of performance and power consumption, and limits the ability for storage nodes to perform sophisticated optimizations.

Innovation Solution

Incorporating a processing unit within a solid state drive (SSD) capable of performing operations like encryption, decryption, compression, decompression, and in-line indexing, allowing data to be processed within the SSD without transferring it to the server CPU, using a digital signal processing core and microcontroller integrated on a single semiconductor chip.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If data is read from storage node, processed by server CPU, and stored back, then data processing operations can be performed, but power consumption and processing time increase significantly

Engineering Contradiction:

Improvedata processing throughputVSAvoidpower consumption

Core Design Contradiction:

ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent combines storage and processing functions into a single storage node by integrating a processing unit directly into the storage device. This merging eliminates the need to transfer data between storage node and server CPU, allowing processing operations to be performed where data resides, thereby reducing power consumption while maintaining processing throughput.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The storage node performs data processing operations autonomously using its integrated processing unit without requiring server CPU intervention. The storage node serves itself by executing search, analysis, encryption, and other data operations locally, reducing the energy burden on the server while maintaining high productivity.

Inventive Principle:

Principle #25Self-service

2Productivity

If server CPU processes data from storage node, then data operations can be performed, but server resources are occupied and other user operations are blocked

Engineering Contradiction:

Improvedata processing capabilityVSAvoidserver availability for other operations

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent segments processing tasks from the server CPU and relocates them to the storage node's integrated processing unit. This segmentation allows the server to dedicate its resources to other user operations while the storage node independently handles data processing, search, and analysis tasks, improving overall system availability and ease of operation.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The storage node autonomously performs data processing operations without requiring server CPU resources. By enabling the storage node to serve its own processing needs, the server remains available for other operations, resolving the contradiction between processing capability and server availability.

Inventive Principle:

Principle #25Self-service

3Productivity

If data is transferred from storage node to server CPU, then processing can occur, but communication bus is tied up and performance decreases

Engineering Contradiction:

Improvedata processing speedVSAvoiddata transfer time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent merges the processing unit with the storage node, eliminating the need for data transfer over the communication bus. Processing operations are performed in-place at the storage node, completely removing data transfer time from the processing pipeline and maximizing processing speed without bus contention.

Inventive Principle:

Principle #5Merging (Combining)

Solution Approach 2:

The integrated processing unit acts as an intermediary between storage memory and the external server, performing all necessary data operations locally. This eliminates the need for data to traverse the communication bus to and from the server CPU, removing transfer delays and improving overall processing speed.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Adaptability or versatility

If standard storage node interface is used, then compatibility is maintained, but storage node cannot perform data operations independently

Engineering Contradiction:

Improveinterface compatibilityVSAvoidin-node processing capability

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent implements a universal storage node architecture that maintains compatibility with standard interfaces (SATA, Fibre Channel, SAS) while simultaneously enabling independent data processing capabilities. The storage node can function both as a traditional storage device and as an autonomous processing unit, adapting to different operational modes without sacrificing compatibility or processing capability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS9021189B2System and method for performing efficient processing of data stored in a storage node

Publication Date: 2015.04.28 NETLIST INC

AI summary

A system and method of providing enhanced data processing and analysis in a storage device, such as a solid state drive (SSD). The SSD includes flash memory and an SSD processing unit capable of executing searches on the data in the flash memory without returning all of the data to be searched to a host CPU outside of the SSD. Other processing capabilities incorporated into the SSD may include encryption and decryption, compression and decompression, and in-line indexing of data.