Reducing Latency in Virtual Machines Enabling Tactile Internet for Human Machine Co-working

Zuo Xiang, Frank Gabriel, Elena Urbano, Giang T. Nguyen, Martin Reisslein, Frank H.P. Fitzek

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Software Defined Networking (SDN) and Network Function Virtualization (NFV) processed in Multi-access Edge Computing (MEC) cloud systems have been proposed as critical paradigms for achieving the low latency requirements of the tactile Internet. While Virtual Network Functions (VNFs) allow greater flexibility compared to hardware based solutions, the VNF abstraction also introduces additional packet processing delays. In this paper, we investigate the practical feasibility of NFV with respect to the tactile Internet latency requirements. We develop, implement, and evaluate Chain bAsed Low latency VNF ImplemeNtation (CALVIN), a low-latency management framework for distributed Service Function Chains (SFCs). CALVIN classifies VNFs into elementary, basic, and advanced VNFs. CALVIN implements elementary and basic VNFs in the kernel space, while advanced VNFs are implemented in the user space. Throughout, CALVIN employs a distributed mapping with one VNF per Virtual Machine (VM) in a MEC system. Moreover, CALVIN avoids the metadata structure processing and batch processing of packets in the conventional Linux networking stack so as to achieve short per-packet latencies. Our rigorous measurements on off-the-shelf conventional networking and computing hardware demonstrate that CALVIN achieves round-trip times from a MEC ingress point via two elementary forwarding VNFs (one in kernel space and one in user space) and a MEC server to a MEC egress point on the order of 0.32 ms. Our measurements also indicate that MEC network coding and encryption are feasible for small 256 byte packets with an MEC latency budget of 0.35 ms; whereas, large 1400 byte packets can complete the network coding, but not the encryption within the 0.35 ms.

Original languageEnglish (US)
JournalIEEE Journal on Selected Areas in Communications
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Internet
Network coding
Cryptography
Virtual machine
Hardware
Cloud computing
Processing
Metadata
Servers

Keywords

  • Bridges
  • Cloud computing
  • Communication networks
  • Delays
  • Network function virtualization
  • Servers

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Electrical and Electronic Engineering

Cite this

Reducing Latency in Virtual Machines Enabling Tactile Internet for Human Machine Co-working. / Xiang, Zuo; Gabriel, Frank; Urbano, Elena; Nguyen, Giang T.; Reisslein, Martin; Fitzek, Frank H.P.

In: IEEE Journal on Selected Areas in Communications, 01.01.2019.

Research output: Contribution to journalArticle

@article{e324f5fb91954a9f8ef4c217dffa4b71,
title = "Reducing Latency in Virtual Machines Enabling Tactile Internet for Human Machine Co-working",
abstract = "Software Defined Networking (SDN) and Network Function Virtualization (NFV) processed in Multi-access Edge Computing (MEC) cloud systems have been proposed as critical paradigms for achieving the low latency requirements of the tactile Internet. While Virtual Network Functions (VNFs) allow greater flexibility compared to hardware based solutions, the VNF abstraction also introduces additional packet processing delays. In this paper, we investigate the practical feasibility of NFV with respect to the tactile Internet latency requirements. We develop, implement, and evaluate Chain bAsed Low latency VNF ImplemeNtation (CALVIN), a low-latency management framework for distributed Service Function Chains (SFCs). CALVIN classifies VNFs into elementary, basic, and advanced VNFs. CALVIN implements elementary and basic VNFs in the kernel space, while advanced VNFs are implemented in the user space. Throughout, CALVIN employs a distributed mapping with one VNF per Virtual Machine (VM) in a MEC system. Moreover, CALVIN avoids the metadata structure processing and batch processing of packets in the conventional Linux networking stack so as to achieve short per-packet latencies. Our rigorous measurements on off-the-shelf conventional networking and computing hardware demonstrate that CALVIN achieves round-trip times from a MEC ingress point via two elementary forwarding VNFs (one in kernel space and one in user space) and a MEC server to a MEC egress point on the order of 0.32 ms. Our measurements also indicate that MEC network coding and encryption are feasible for small 256 byte packets with an MEC latency budget of 0.35 ms; whereas, large 1400 byte packets can complete the network coding, but not the encryption within the 0.35 ms.",
keywords = "Bridges, Cloud computing, Communication networks, Delays, Network function virtualization, Servers",
author = "Zuo Xiang and Frank Gabriel and Elena Urbano and Nguyen, {Giang T.} and Martin Reisslein and Fitzek, {Frank H.P.}",
year = "2019",
month = "1",
day = "1",
doi = "10.1109/JSAC.2019.2906788",
language = "English (US)",
journal = "IEEE Journal on Selected Areas in Communications",
issn = "0733-8716",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Reducing Latency in Virtual Machines Enabling Tactile Internet for Human Machine Co-working

AU - Xiang, Zuo

AU - Gabriel, Frank

AU - Urbano, Elena

AU - Nguyen, Giang T.

AU - Reisslein, Martin

AU - Fitzek, Frank H.P.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - Software Defined Networking (SDN) and Network Function Virtualization (NFV) processed in Multi-access Edge Computing (MEC) cloud systems have been proposed as critical paradigms for achieving the low latency requirements of the tactile Internet. While Virtual Network Functions (VNFs) allow greater flexibility compared to hardware based solutions, the VNF abstraction also introduces additional packet processing delays. In this paper, we investigate the practical feasibility of NFV with respect to the tactile Internet latency requirements. We develop, implement, and evaluate Chain bAsed Low latency VNF ImplemeNtation (CALVIN), a low-latency management framework for distributed Service Function Chains (SFCs). CALVIN classifies VNFs into elementary, basic, and advanced VNFs. CALVIN implements elementary and basic VNFs in the kernel space, while advanced VNFs are implemented in the user space. Throughout, CALVIN employs a distributed mapping with one VNF per Virtual Machine (VM) in a MEC system. Moreover, CALVIN avoids the metadata structure processing and batch processing of packets in the conventional Linux networking stack so as to achieve short per-packet latencies. Our rigorous measurements on off-the-shelf conventional networking and computing hardware demonstrate that CALVIN achieves round-trip times from a MEC ingress point via two elementary forwarding VNFs (one in kernel space and one in user space) and a MEC server to a MEC egress point on the order of 0.32 ms. Our measurements also indicate that MEC network coding and encryption are feasible for small 256 byte packets with an MEC latency budget of 0.35 ms; whereas, large 1400 byte packets can complete the network coding, but not the encryption within the 0.35 ms.

AB - Software Defined Networking (SDN) and Network Function Virtualization (NFV) processed in Multi-access Edge Computing (MEC) cloud systems have been proposed as critical paradigms for achieving the low latency requirements of the tactile Internet. While Virtual Network Functions (VNFs) allow greater flexibility compared to hardware based solutions, the VNF abstraction also introduces additional packet processing delays. In this paper, we investigate the practical feasibility of NFV with respect to the tactile Internet latency requirements. We develop, implement, and evaluate Chain bAsed Low latency VNF ImplemeNtation (CALVIN), a low-latency management framework for distributed Service Function Chains (SFCs). CALVIN classifies VNFs into elementary, basic, and advanced VNFs. CALVIN implements elementary and basic VNFs in the kernel space, while advanced VNFs are implemented in the user space. Throughout, CALVIN employs a distributed mapping with one VNF per Virtual Machine (VM) in a MEC system. Moreover, CALVIN avoids the metadata structure processing and batch processing of packets in the conventional Linux networking stack so as to achieve short per-packet latencies. Our rigorous measurements on off-the-shelf conventional networking and computing hardware demonstrate that CALVIN achieves round-trip times from a MEC ingress point via two elementary forwarding VNFs (one in kernel space and one in user space) and a MEC server to a MEC egress point on the order of 0.32 ms. Our measurements also indicate that MEC network coding and encryption are feasible for small 256 byte packets with an MEC latency budget of 0.35 ms; whereas, large 1400 byte packets can complete the network coding, but not the encryption within the 0.35 ms.

KW - Bridges

KW - Cloud computing

KW - Communication networks

KW - Delays

KW - Network function virtualization

KW - Servers

UR - http://www.scopus.com/inward/record.url?scp=85063097866&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063097866&partnerID=8YFLogxK

U2 - 10.1109/JSAC.2019.2906788

DO - 10.1109/JSAC.2019.2906788

M3 - Article

AN - SCOPUS:85063097866

JO - IEEE Journal on Selected Areas in Communications

JF - IEEE Journal on Selected Areas in Communications

SN - 0733-8716

ER -