Memory resource management in VMware ESX server
ACM SIGOPS Operating Systems Review - OSDI '02: Proceedings of the 5th symposium on Operating systems design and implementation
AMP: adaptive multi-stream prefetching in a shared cache
FAST '07 Proceedings of the 5th USENIX conference on File and Storage Technologies
Competitive prefetching for concurrent sequential I/O
Proceedings of the 2nd ACM SIGOPS/EuroSys European Conference on Computer Systems 2007
On the design of a new Linux readahead framework
ACM SIGOPS Operating Systems Review - Research and developments in the Linux kernel
AFIPS '74 Proceedings of the May 6-10, 1974, national computer conference and exposition
Memory resource allocation for file system prefetching: from a supply chain management perspective
Proceedings of the 4th ACM European conference on Computer systems
SmartVisor: towards an efficient and compatible virtualization platform for embedded system
Proceedings of the Second Workshop on Isolation and Integration in Embedded Systems
Hi-index | 0.00 |
In a virtualized system, the hypervisor may be forced to reclaim memory by swapping out pages of guest operating systems (OSes) when the regular memory balancing mechanisms, such as page sharing and ballooning, fail to revoke enough memory for reallocation purpose, which always leads to serious performance degradation. In this paper, we introduce Adaptive Swap Prefetcher (ASP) and Host Swapping Notifier (HSN), the effective and lightweight solutions to gracefully reduce the degradation in system performance when host swapping is triggered. ASP smartly prefetches more pages from the host swap file as long as the good spatial locality persists so as to reduce disk transfers. The guest OS will be notified by HSN when the hypervisor evicts pages, which then hides those pages from its memory reclamation routines to eliminate unnecessary guest swapping and to prevent the occurrence of double paging anomaly. Currently ASP and HSN are implemented in KVM, experimental results show that guest performance can be improved by a factory of 1.4x and 2x respectively using ASP and HSN.