Integrated Server Sled Storage Density and Reliability

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing server designs with centralized controllers lead to inefficiencies in space utilization and reliability, as they often leave unused space in server chassis, and a single component failure can render all hard disk drives inaccessible, reducing system reliability and storage capacity.

Innovation Solution

Integrating a server device within the same chassis as mass storage devices, allowing it to occupy equivalent volume space, thereby eliminating wasted space and creating independent logical nodes that can continue operations even if one component fails, enhancing storage density and reliability.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If standard off-the-shelf hard disk drives are used in server designs, then cost effectiveness is improved, but space utilization deteriorates due to substantial wasted space in server chassis

Engineering Contradiction:
Improvecost effectivenessVSAvoidspace utilization
Core Design Contradiction:
Ease of manufactureVSVolume of moving object

Solution Approach 1:

The patent merges the controller and hard disk drives into a single integrated unit that fits within the server chassis. This integration eliminates the wasted space that occurs when separate controllers and drives are used, while still allowing the use of standard off-the-shelf hard disk drives. The integrated unit combines multiple components into one compact assembly that maximizes space utilization in the server.

Inventive Principle:
Principle #5Merging (Combining)

2Device complexity

If a centralized controller is used to control accesses to hard disk drives, then device complexity is reduced, but reliability deteriorates because a controller failure causes all hard disk drives to be unreachable

Engineering Contradiction:
Improvecontroller architectureVSAvoidsystem availability
Core Design Contradiction:
Device complexityVSReliability

Solution Approach 1:

The patent segments the storage system into multiple independent units, each with its own controller and hard disk drives. This segmentation means that if one controller fails, only the hard disk drives in that specific unit are affected, while other units continue to operate normally. This eliminates the single point of failure problem associated with centralized controllers, thereby improving reliability without significantly increasing overall system complexity.

Inventive Principle:
Principle #1Segmentation

3Productivity

If rack systems include multiple rack-mounted components to increase computing capacity, then productivity is improved, but heat generation increases reaching up to 10 kilowatts

Engineering Contradiction:
Improvecomputing capacityVSAvoidwaste heat
Core Design Contradiction:
ProductivityVSLoss of energy

Solution Approach 1:

The patent merges multiple components (controller and hard disk drives) into integrated units that are more space-efficient and potentially more thermally efficient. By consolidating components into compact integrated units rather than having them as separate rack-mounted components, the system can achieve similar computing capacity with reduced heat generation per unit of capacity.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS11395433B2Server system
Publication Date: 2022.07.19 AMAZON TECH INC
  • US11395433B2 patent drawing
  • US11395433B2 patent drawing
  • US11395433B2 patent drawing

AI summary

A system includes a rack and one or more server systems mounted in the rack. A server system includes a chassis with one or more arrays of devices in the chassis. Each array includes mass storage devices and a server device mounted in the array within the chassis of the server system. The server device occupies no more than an equivalent volume of space in the chassis as one of the mass storage devices. A set of mass storage devices of an array and the server device of the array form a logical node, and a server system may include multiple logical nodes in the same chassis. Each array of devices may be located in a sled that can move into and out of the chassis of the server system.