RAID Controller Segmented XOR Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional RAID systems face inefficiencies in performing XOR operations due to the need for data to be completely written into system memory before performing parity updates, leading to excessive waiting time and reduced performance.

Innovation Solution

A computer system with a RAID control method that includes a data length detecting unit to trigger a fault-tolerant data computing operation when a unitary length of data is reached, allowing for timely XOR operations without waiting for complete data transfer, utilizing a hard disk controller and a fault-tolerant data computing unit within the core logic device or RAID card.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is completely written into system memory before performing XOR operations, then data integrity is ensured, but waiting time increases and performance decreases

Engineering Contradiction:
Improvedata integrityVSAvoidwaiting time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by initiating XOR operations on data segments before the entire data transfer is complete. The controller performs fault-tolerant computing on received data segments immediately when they arrive, rather than waiting for all data to be transferred to system memory. This allows parity data to be calculated in advance for segments that have already been received, reducing the overall waiting time while maintaining data integrity through proper segment management.

Inventive Principle:
Principle #10Preliminary action

2Manufacturing precision

If data is completely transferred to system memory before XOR operations, then processing accuracy is maintained, but system performance deteriorates

Engineering Contradiction:
Improveprocessing accuracyVSAvoidsystem performance
Core Design Contradiction:
Manufacturing precisionVSProductivity

Solution Approach 1:

The patent applies segmentation by dividing the data transfer and processing into multiple independent segments. Instead of treating the entire data transfer as a single unit that must complete before processing, the controller processes data in segments as they arrive. Each segment can be independently processed for XOR operations, allowing parallel processing and improving system performance while maintaining processing accuracy through proper segment boundary management.

Inventive Principle:
Principle #1Segmentation

3Productivity

If the RAID control function is integrated into the core logic device, then data transfer efficiency improves, but device complexity increases

Engineering Contradiction:
Improvedata transfer efficiencyVSAvoiddevice complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies merging by integrating the RAID control function directly into the core logic device, combining storage control and processing functions into a single unified device. This integration eliminates the need for separate communication interfaces and data transfer paths between independent RAID controllers and storage devices, thereby improving data transfer efficiency. The increased device complexity is accepted as a trade-off for achieving the performance benefits of integrated control.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS7757130B2Computer system having raid control function and raid control method
Publication Date: 2010.07.13 VIA TECH INC
  • US7757130B2 patent drawing
  • US7757130B2 patent drawing
  • US7757130B2 patent drawing

AI summary

RAID control of multiple hard disk drives in a computer system includes performing a fault-tolerant data computing operation for a written data. The timing for performing the fault-tolerant data computing operation is determined by accessing a data stored in one of the hard disk drives, detecting a partial data length of a data stream having been transmitted from the hard disk drive to the computer system, issuing a triggering signal when the data length has reached a unitary length less than the total length of the data stream, and then performing the fault-tolerant data computing operation with the unitary length of data in response to the triggering signal.