Memory Page Merging for Virtualization Cache Pollution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional virtual computing environments face cache pressure issues due to guest and host machines swapping in and out of system control, leading to cache pollution, which is not effectively mitigated by increasing cache size, and existing techniques do not address the challenge of shared caches between guest and host machines.
Innovation Solution
The system matches guest machines with host machines having common operating system instructions, merges memory pages, and optimizes instruction sets to share whole memory pages, reducing cache pressure and improving cache locality by using memory page comparison logic and arrangement logic to align instructions at page boundaries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If cache size is increased to mitigate cache pressure, then cache capacity is improved, but device complexity and cost increase
Solution Approach 1:
The patent merges memory pages between guest and host machines by identifying identical pages and mapping them to shared physical memory locations. This consolidation eliminates duplicate data storage, reducing the total cache capacity needed while maintaining system functionality.
Solution Approach 2:
The patent creates a universal memory space where physical memory pages serve multiple purposes - simultaneously satisfying both guest machine virtual memory requests and host machine memory needs when content is identical. This multi-functionality reduces overall memory requirements.
2Speed
If guest and host machines share cache space, then memory access time is improved, but cache pollution increases
Solution Approach 1:
The patent merges identical memory pages from guest and host into shared physical memory pages, ensuring that both machines access the same physical cache lines. This eliminates cache pollution caused by duplicate copies while maintaining fast access times through unified caching.
Solution Approach 2:
The patent applies content-based sharing where memory pages are shared only when their content is identical, creating localized sharing relationships. This selective approach ensures that shared pages benefit from caching while avoiding pollution from non-identical content.
3Quantity of substance
If memory pages are merged between guest and host, then cache pressure is reduced, but memory management complexity increases
Solution Approach 1:
The patent performs preliminary comparison of memory page content between guest and host machines to identify identical pages before establishing shared mappings. This advance preparation simplifies subsequent memory management by pre-identifying sharing candidates.
Solution Approach 2:
The patent uses content copying and comparison techniques to identify identical memory pages. By comparing page content and detecting duplicates, the system can establish shared mappings without complex analysis, reducing management overhead.
4Speed
If instruction sets are optimized for sharing, then cache locality is improved, but processing time for optimization increases
Solution Approach 1:
The patent segments memory management into discrete memory page units, comparing and sharing at the page level rather than individual instructions. This segmentation reduces optimization complexity and processing time while still achieving improved cache locality through page-level sharing.
Solution Approach 2:
The patent applies partial optimization by focusing on memory pages that are likely to be shared based on content similarity, rather than optimizing all instructions. This selective approach reduces processing time while achieving sufficient cache locality improvement.
Data Source
AI summary
Systems and techniques for memory and resource management in a virtual computing environment are disclosed herein. For example, in some embodiments, an apparatus for memory management in a virtual computing environment may include: a storage device; memory page comparison logic, coupled to the storage device, to determine that a first memory page of instructions, stored in the storage device, for a guest machine in the virtual computing environment is identical to a second memory page of instructions, stored in the storage device, for a host machine in the virtual computing environment, wherein the guest machine is hosted by the host machine; and merge logic, coupled to the memory page comparison logic, to, in response to a determination that the first memory page is identical to the second memory page, map the first memory page to the second memory page. Other embodiments may be disclosed and/or claimed.