.png){kind=icon}
.png){kind=icon}
Return
Serverless Computing for QoS-Effective NFV in the Cloud Edge
.png){kind=icon}
PRE
AI
DOI：10.1109/MCOM.001.2300182.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Network Softwarization, particularly Network Function Virtualization (NFV), is revolutionizing networking by separating the hardware from the supported network functions, demonstrating unprecedented flexibility and cost-efficiency. With the rise of edge computing, NFV has found a strategic application at the network edge, where network functions are deployed closer to end-users and devices with improved performance and reduced latency. In this context, the serverless paradigm is gaining attention due to its fine-grained scalability, reduced management efforts, and resource efficiency. Despite the advantages, challenges arise from the ephemeral nature of serverless functions, leading to latency issues and hampering the creation of efficient coordination and orchestration mechanisms. Adopting this paradigm in resource-constrained scenarios, such as cloud edges, is even more challenging due to the insufficient presence of proper quality-of-service (QoS) mechanisms in these platforms. This article introduces our novel time effective middleware for priority oriented serverless network function virtualization (TEMPOS4NFV), a QoS-aware platform specialized in hosting serverless network functions in resource-constrained environments. TEMPOS4NFV addresses resource contention and coordination challenges, offering effective end-to-end QoS differentiation for network workloads executed over multiple federated sites. In addition, this article examines TEMPOS4NFV hosting a real virtual private network (VPN) case, demonstrating its ability to execute concurrent multi-site QoS differentiated workloads.
Keywords:
Quality of service
Method of moments
Scalability
Virtual private networks
Computer architecture
Optimization
Serverless computing
Network function virtualization
Cloud computing
Journal
IF:
8.2
Papers:
6.9K
Citations:
2.2W