Swift Birth and Quick DeathCitation formats

  • External authors:
  • Vlad Nitu
  • Alain Tchana
  • Daniel Chiba
  • Antonio Barbalace
  • Daniel Hagimont
  • Binoy Ravindran

Standard

Swift Birth and Quick Death : Enabling Fast Parallel Guest Boot and Destruction in the Xen Hypervisor. / Nitu, Vlad; Olivier, Pierre; Tchana, Alain; Chiba, Daniel; Barbalace, Antonio; Hagimont, Daniel; Ravindran, Binoy.

In: ACM SIGPLAN Notices, 01.04.2017, p. 1-14.

Research output: Contribution to journal › Article › peer-review

Harvard

Nitu, V, Olivier, P, Tchana, A, Chiba, D, Barbalace, A, Hagimont, D & Ravindran, B 2017, 'Swift Birth and Quick Death: Enabling Fast Parallel Guest Boot and Destruction in the Xen Hypervisor', ACM SIGPLAN Notices, pp. 1-14. https://doi.org/10.1145/3050748.3050758

APA

Nitu, V., Olivier, P., Tchana, A., Chiba, D., Barbalace, A., Hagimont, D., & Ravindran, B. (2017). Swift Birth and Quick Death: Enabling Fast Parallel Guest Boot and Destruction in the Xen Hypervisor. ACM SIGPLAN Notices, 1-14. https://doi.org/10.1145/3050748.3050758

Vancouver

Nitu V, Olivier P, Tchana A, Chiba D, Barbalace A, Hagimont D et al. Swift Birth and Quick Death: Enabling Fast Parallel Guest Boot and Destruction in the Xen Hypervisor. ACM SIGPLAN Notices. 2017 Apr 1;1-14. https://doi.org/10.1145/3050748.3050758

Author

Nitu, Vlad ; Olivier, Pierre ; Tchana, Alain ; Chiba, Daniel ; Barbalace, Antonio ; Hagimont, Daniel ; Ravindran, Binoy. / Swift Birth and Quick Death : Enabling Fast Parallel Guest Boot and Destruction in the Xen Hypervisor. In: ACM SIGPLAN Notices. 2017 ; pp. 1-14.

Bibtex

@article{0c430c52c2ed435cbd8bac016b61b42b,
title = "Swift Birth and Quick Death: Enabling Fast Parallel Guest Boot and Destruction in the Xen Hypervisor",
abstract = "The ability to quickly set up and tear down a virtual machine is critical for today's cloud elasticity, as well as in numerous other scenarios: guest migration/consolidation, event-driven invocation of micro-services, dynamically adaptive unikernel-based applications, micro-reboots for security or stability, etc.In this paper, we focus on the process of setting up/freeing the hypervisor and host control layer data structures at boot/destruction time, showing that it does not scale in current virtualization solutions. In addition to the direct overhead of long VM set-up/destruction times, we demonstrate by experimentation the indirect costs on real world auto scaling systems. Focusing on the popular Xen hypervisor, we identify three critical issues hindering the scalability of the boot and destruction processes: serialized boot, unscalable interactions with the Xenstore at guest creation time, and remote NUMA memory scrubbing at destruction time. For each of these issues we present the design and implementation of a solution in the Xen infrastructure: parallel boot with fine-grained locking, caching of Xenstore data, and local NUMA scrubbing. We evaluate these solutions using micro-benchmarks, macro-benchmarks, and real world datacenter traces. Results show that our work improves the current Xen implementation by a significant factor, for example macro-benchmarks indicate a speedup of more than 4X in high-load scenarios.",
author = "Vlad Nitu and Pierre Olivier and Alain Tchana and Daniel Chiba and Antonio Barbalace and Daniel Hagimont and Binoy Ravindran",
year = "2017",
month = apr,
day = "1",
doi = "10.1145/3050748.3050758",
language = "English",
pages = "1--14",
journal = "ACM SIGPLAN Notices",
issn = "0362-1340",
publisher = "ACM Digital Library",
note = "The 13th ACM SIGPLAN/SIGOPS International Conference ; Conference date: 08-04-2017 Through 09-04-2017",

}

RIS

TY - JOUR

T1 - Swift Birth and Quick Death

T2 - The 13th ACM SIGPLAN/SIGOPS International Conference

AU - Nitu, Vlad

AU - Olivier, Pierre

AU - Tchana, Alain

AU - Chiba, Daniel

AU - Barbalace, Antonio

AU - Hagimont, Daniel

AU - Ravindran, Binoy

PY - 2017/4/1

Y1 - 2017/4/1

N2 - The ability to quickly set up and tear down a virtual machine is critical for today's cloud elasticity, as well as in numerous other scenarios: guest migration/consolidation, event-driven invocation of micro-services, dynamically adaptive unikernel-based applications, micro-reboots for security or stability, etc.In this paper, we focus on the process of setting up/freeing the hypervisor and host control layer data structures at boot/destruction time, showing that it does not scale in current virtualization solutions. In addition to the direct overhead of long VM set-up/destruction times, we demonstrate by experimentation the indirect costs on real world auto scaling systems. Focusing on the popular Xen hypervisor, we identify three critical issues hindering the scalability of the boot and destruction processes: serialized boot, unscalable interactions with the Xenstore at guest creation time, and remote NUMA memory scrubbing at destruction time. For each of these issues we present the design and implementation of a solution in the Xen infrastructure: parallel boot with fine-grained locking, caching of Xenstore data, and local NUMA scrubbing. We evaluate these solutions using micro-benchmarks, macro-benchmarks, and real world datacenter traces. Results show that our work improves the current Xen implementation by a significant factor, for example macro-benchmarks indicate a speedup of more than 4X in high-load scenarios.

AB - The ability to quickly set up and tear down a virtual machine is critical for today's cloud elasticity, as well as in numerous other scenarios: guest migration/consolidation, event-driven invocation of micro-services, dynamically adaptive unikernel-based applications, micro-reboots for security or stability, etc.In this paper, we focus on the process of setting up/freeing the hypervisor and host control layer data structures at boot/destruction time, showing that it does not scale in current virtualization solutions. In addition to the direct overhead of long VM set-up/destruction times, we demonstrate by experimentation the indirect costs on real world auto scaling systems. Focusing on the popular Xen hypervisor, we identify three critical issues hindering the scalability of the boot and destruction processes: serialized boot, unscalable interactions with the Xenstore at guest creation time, and remote NUMA memory scrubbing at destruction time. For each of these issues we present the design and implementation of a solution in the Xen infrastructure: parallel boot with fine-grained locking, caching of Xenstore data, and local NUMA scrubbing. We evaluate these solutions using micro-benchmarks, macro-benchmarks, and real world datacenter traces. Results show that our work improves the current Xen implementation by a significant factor, for example macro-benchmarks indicate a speedup of more than 4X in high-load scenarios.

U2 - 10.1145/3050748.3050758

DO - 10.1145/3050748.3050758

M3 - Article

SP - 1

EP - 14

JO - ACM SIGPLAN Notices

JF - ACM SIGPLAN Notices

SN - 0362-1340

Y2 - 8 April 2017 through 9 April 2017

ER -