Serverless Storage Management via OS Offloading and NVRAM Buffering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Traditional storage systems face inefficiencies in data management and storage operations, particularly in handling flash storage systems where lower-level processes are often required, leading to unnecessary write operations and reduced reliability.

Innovation Solution

Implementing a direct-mapped flash storage system where the operating system initiates and controls processes, offloading device management responsibilities from storage drives, and utilizing non-volatile random access memory (NVRAM) to improve write latency and reliability by buffering data before writing it to storage drives.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional storage systems use lower-level processes for flash storage management, then device control is maintained, but redundant write operations increase and reliability decreases

Engineering Contradiction:
Improvestorage system reliabilityVSAvoidwrite operation efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent extracts device management responsibilities from the storage drive itself and transfers them to the operating system. The storage drive presents a simple block device interface while the OS handles higher-level management tasks including write buffering, caching, and coordination between multiple storage devices. This extraction eliminates redundant write operations and improves reliability by centralizing control logic.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an intermediary layer (the operating system) between the application and the storage drive. This intermediary manages write operations through buffering and caching mechanisms, coordinating writes across multiple storage devices to prevent redundant operations. The intermediary layer abstracts the complexity from individual drives while improving overall system reliability and write efficiency.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of time

If NVRAM is used for buffering data before writing to storage drives, then write latency is improved and reliability increases, but system complexity increases

Engineering Contradiction:
Improvewrite latencyVSAvoidstorage system complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent implements preliminary action by using NVRAM to buffer write operations before they are committed to permanent storage. Data is first written to the fast NVRAM medium, allowing the system to acknowledge writes immediately without waiting for slower storage drive operations. This preliminary buffering significantly reduces write latency while the coordination logic manages the complexity of synchronizing data across multiple storage devices.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS12056396B2Storage-aware management for serverless functions
Publication Date: 2024.08.06 PURE STORAGE INC
  • US12056396B2 patent drawing
  • US12056396B2 patent drawing
  • US12056396B2 patent drawing

AI summary

An illustrative method includes a storage-aware serverless function management system receiving a request to execute a serverless function instance of a serverless function implemented in a serverless system, the serverless function instance associated with a component of a storage system, determining a portion of the component accessible to the serverless function instance based on a storage system policy associated with the storage system, and executing the serverless function instance using the portion of the component of the storage system.