Litcius/Paper detail

PTEMagnet: fine-grained physical memory reservation for faster page walks in public clouds

Artemiy Margaritov, Dmitrii Ustiugov, Amna Shahab, Boris Grot

202118 citationsDOIOpen Access PDF

Abstract

The last few years have seen a rapid adoption of cloud computing for data-intensive tasks. In the cloud environment, it is common for applications to run under virtualization and to share a virtual machine with other applications (e.g., in a virtual private cloud setup). In this setting, our work identifies a new address translation bottleneck caused by memory fragmentation stemming from the interaction of virtualization, colocation, and the Linux memory allocator. The fragmentation results in the effective cache footprint of the host PT being larger than that of the guest PT. The bloated footprint of the host PT leads to frequent cache misses during nested page walks, increasing page walk latency.

Topics & Concepts

Computer scienceVirtualizationCloud computingOperating systemTranslation lookaside bufferCacheBottleneckVirtual machineMemory footprintPage cacheLatency (audio)Virtual memoryParallel computingCPU cacheDistributed computingMemory managementCache algorithmsPhysical addressEmbedded systemOverlayTelecommunicationsCloud Computing and Resource ManagementAdvanced Data Storage TechnologiesAdvanced Neural Network Applications
PTEMagnet: fine-grained physical memory reservation for faster page walks in public clouds | Litcius