Docker Container Deployment in Distributed Fog Infrastructures with Checkpoint/Restart

Abstract : In fog computing environments container deployment is a frequent operation which often lies in the critical path of services being delivered to an end user. Although creating a container can be very fast, the container's application needs to start before the container starts producing useful work. Depending on the application this startup process can be arbitrarily long. To speed up the application startup times we propose to snapshot the state of fully-deployed containers and restart future container instances from a pre-started application state. In our evaluations based on 14 real micro-service containers, this technique effectively reduces the startup phase with speedups between 1x (no speedup) and 60x.
Complete list of metadatas

Cited literature [34 references]  Display  Hide  Download

https://hal.inria.fr/hal-02473333
Contributor : Guillaume Pierre <>
Submitted on : Monday, February 10, 2020 - 4:29:12 PM
Last modification on : Tuesday, February 18, 2020 - 10:27:41 AM

File

main.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02473333, version 1

Citation

Arif Ahmed, Apoorve Mohan, Gene Cooperman, Guillaume Pierre. Docker Container Deployment in Distributed Fog Infrastructures with Checkpoint/Restart. IEEE International Conference on Mobile Cloud Computing, Apr 2020, Oxford, United Kingdom. ⟨hal-02473333⟩

Share

Metrics

Record views

49

Files downloads

65