Write a Blog >>
VEE 2017
Sat 8 - Sun 9 April 2017 Xi'an, China
Sat 8 Apr 2017 16:35 - 17:00 at Zhu Que Room - Introspection Chair(s): Paolo Bonzini

We introduce Sky, an extension to the VMM that gathers insights and information by intercepting system calls made by guest applications. We show how Sky gains three specific insights – guest file-size information, metadata-data distinction, and file-content hints – and uses said information to enhance virtualized-storage performance. By caching small files and metadata with higher priority, Sky reduces the runtime by 2.3 to 8.8 times for certain workloads. Sky also achieves 4.5 to 18.7 times reduction in the runtime of an open-source block-layer deduplication system by exploiting hints about file contents. Sky works underneath both Linux and FreeBSD guests, as well as under a range of file systems, thus enabling portable and general VMM-level optimization underneath a wide range of storage stacks.

this URL might only work when visiting from a https://doi.org/10.1145/3050748.3050755 URL.
Sky Presentation Slides (vee17_Sky_presentation_slides.pdf)850KiB

Conference Day
Sat 8 Apr

Displayed time zone: Azores change

15:45 - 17:00
IntrospectionSession 3 at Zhu Que Room
Chair(s): Paolo BonziniRed Hat, Inc.
Dancing with Wolves: Towards Practical Event-driven VMM Monitoring
Session 3
Liang DengState Key Laboratory for Novel Software Technology, Nanjing University; SANGFOR TECHNOLOGIES CO., LTD., Peng LiuPurdue University, Jun XuInformation Sciences and Technology, Pennsylvania State University, Ping ChenInformation Sciences and Technology, Pennsylvania State University, Qingkai ZengState Key Laboratory for Novel Software Technology, Nanjing University, China
Safe Inspection of Live Virtual Machines
Session 3
Sahil SunejaIBM Research, Ricardo KollerIBM Research, Canturk IsciIBM Research, Eyal de LaraUniversity of Toronto, Ali HashemiUniversity of Toronto, Arnamoy Bhattacharyya, Cristiana AmzaUniversity of Toronto
Improving Virtualized Storage Performance with Sky
Session 3
DOI Authorizer link File Attached