Virtualized Server Data De-duplication via Pointer Replacement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Virtualized server environments store significant amounts of redundant data in memory and storage, leading to inefficient use of resources, as multiple identical operating systems and data instances are maintained separately, resulting in unnecessary storage of redundant data.
Innovation Solution
A data de-duplication application is implemented within a virtualized server environment using a virtualization layer to identify and replace redundant data with pointers or references to a single instance, allowing for efficient storage and processing by decoupling physical hardware from the operating system and applications, enabling in-line or post-process de-duplication of data in memory and storage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple identical operating systems are maintained separately in memory and storage for each virtual machine, then each virtual machine can operate independently, but significant amounts of redundant data are stored consuming memory and storage space
Solution Approach 1:
The patent merges identical data blocks from multiple virtual machines into a single shared copy in storage. When the same data (e.g., identical OS files) is detected across multiple VMs, the system consolidates these into one master copy and creates references to it, thereby reducing redundant storage while maintaining VM independence through the reference mechanism.
Solution Approach 2:
The patent creates reference copies instead of full data copies. Each virtual machine maintains a reference pointer to the shared data block rather than storing the actual data independently. This allows VMs to access identical data efficiently while the physical storage contains only one instance of the data.
2Adaptability or versatility
If separate operating systems are loaded into memory for each virtual machine, then virtual machines can run in isolation, but redundant instances of data occupy memory space that could be used for other data
Solution Approach 1:
The patent merges identical memory-resident data blocks from multiple virtual machines into a single shared memory allocation. When duplicate data is detected in the memory of different VMs, the system consolidates these into one shared buffer and establishes references, thereby reducing total memory consumption while preserving VM isolation through logical separation.
Solution Approach 2:
The patent creates universal data blocks that can be shared across multiple virtual machines simultaneously. A single data block in memory or storage can serve multiple VMs that require the same data, making the storage resource universal and multi-functional rather than dedicated to a single VM.
3Productivity
If full systems are moved between physical servers, then zero-downtime maintenance is achieved, but significant amounts of data must be transferred and stored
Solution Approach 1:
The patent uses reference copying to accelerate system migration. When a virtual machine is moved between physical servers, instead of transferring all data blocks, the system transfers only the reference pointers to the shared data blocks that already exist in the destination server's storage. This dramatically reduces data transfer volume while maintaining full system functionality.
Data Source
AI summary
A data de-duplication application de-duplicates redundant data on the primary storage read/write pathway of a virtualized server environment. The virtualized server environment comprises one or more server applications operating on a virtualization layer provided on a computer architecture that includes memory (e.g., RAM, cache memory) for temporarily storing data and storage (e.g., disk storage) for persistently storing data. The one or more server applications use the read-write pathway to read data into memory from storage and to write data to storage from memory. The de-duplication application identifies redundant data in memory, storage, or both, and replaces the redundant data with one or more pointers pointing to a single copy of the data. The de-duplication application can operate on fixed or variable size blocks of data and can de-duplicate data either post-process or in-line.