.png){kind=icon}
.png){kind=icon}
Return
Elastic Resource Allocation Algorithms for Collaboration Applications
.png){kind=icon}
PRE
AI
DOI：10.1007/s10922-017-9431-2.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Cloud technologies can provide elasticity to real-time audio and video (A/V) collaboration applications. However, cloud-based collaboration solutions generally operate on a best-effort basis, with neither delivery nor quality guarantees, and high-quality business focused solutions rely on dedicated infrastructure and hardware-based components. This article describes our 2-year of research in the EMD project, which targets to migrate a hardware-based and business focused A/V collaboration solution to a software-based platform hosted in the cloud, providing higher levels of elasticity and reliability. Our focus during this period was an educational collaboration scenario with teachers and students (locally present in the classroom or remotely following the classes). A model of collaboration streaming (e.g. network topology, codecs, stream, streaming workflow, software components) is defined as base for software deployment and preemptive VM allocation techniques. These heuristics are evaluated using a version of the CloudSim simulator extended to generate and simulate realistic collaboration scenarios, to manage network congestion and to monitor a.o. cost and session delay metrics. Our results show that the algorithms reduce costs when compared to previously designed approaches, having an effectiveness of 99% in meeting A/V collaboration setup deadlines, which is a stringent requirement for this collaboration application.
Keywords:
Cloud
Elasticity
Multicast
Media
Education
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
3.9
Papers:
1.0K
Citations:
1.3K