[Paper Review] Enabling Mobility in LTE-Compatible Mobile-edge Computing with Programmable Switches

Ashkan Aghdai, Yang Xu+3 more|arXiv (Cornell University)|May 13, 2019

IoT and Edge/Fog Computing19 references4 citations

TL;DR

This paper proposes the Mobile-Edge Gateway (MEGW), a programmable switch-based solution that enables seamless mobility for mobile-edge computing (MEC) applications in LTE-compatible networks without modifying existing protocols or applications. By leveraging LTE control plane signaling and a two-stage traffic steering mechanism, MEGW transparently redirects user traffic to the nearest edge server, reducing application migrations and simplifying service provisioning in a backward-compatible manner.

ABSTRACT

Network softwarization triggered a new wave of innovation in modern network design. The next generation of mobile networks embraces this trend. Mobile-edge computing (MEC) is a key part of emerging mobile networks that enables ultra-low latency mission-critical application such as vehicle-to vehicle communication. MEC aims at bringing delay-sensitive applications closer to the radio access network to enable ultra-low latency for users and decrease the back-haul pressure on mobile service providers. However, there are no practical solutions to enable mobility at MEC where connections are no longer anchored to the core network and serving applications are supposed to move as their users move. We propose the mobile-edge gateway (MEGW) to address this gap. MEGW enables mobility for MEC applications transparently and without requiring any modifications to existing protocols and applications. MEGW supports mobility by reconstructing mobile users' location via listening to LTE control plane in addition to using two-stage location-dependent traffic steering for edge connections. Networks can incrementally upgrade to support MEC by upgrading some IP router to programmable switches that run MEGW. We have implemented MEGW using P4 language and verified its compatibility with existing LTE networks in a testbed running reference LTE protocol stack. Furthermore, using packet-level simulations we show that the two-stage traffic steering algorithm reduces the number of application migrations and simplifies service provisioning.

Motivation & Objective

• Address the lack of practical solutions for supporting user mobility in mobile-edge computing (MEC) environments where applications must migrate with users to maintain low latency.
• Enable seamless handover of MEC applications across edge servers without requiring modifications to existing LTE protocols, applications, or network components.
• Design a scalable and incremental upgrade path for existing LTE networks to support MEC by upgrading only selected IP routers to programmable switches.
• Implement efficient load balancing across MEC instances using a virtualized cluster abstraction to improve resource utilization and fairness.
• Ensure backward compatibility with existing 4G/LTE control and data planes while enabling full MEC functionality through programmable infrastructure.

Proposed method

• Design and implement MEGW using the P4 programming language on a Netronome NFP4000 programmable switch target.
• Enable transparent location tracking by listening to LTE control plane signaling (e.g., S11/S10 interfaces) to detect user mobility and update edge server mappings.
• Employ a two-stage traffic steering mechanism: first, direct traffic to a regional virtual MEC cluster; second, steer to the optimal physical MEC instance within that cluster.
• Treat multiple small physical MEC clusters as a single large virtual MEC cluster per geographic region to improve load balancing and resource utilization.
• Integrate stateful connection tracking and consistent hashing for load balancing, ensuring connection affinity and minimizing disruption during migrations.
• Verify end-to-end functionality using a testbed with OpenAirInterface’s reference 5G protocol stack and validate compatibility with real LTE control and data planes.

Experimental results

Research questions

• RQ1 How can MEC applications be made mobile in LTE-compatible networks without modifying existing protocols or applications?
• RQ2 What mechanisms enable transparent, low-latency traffic steering to edge servers as users move across the network?
• RQ3 How can a two-stage traffic steering approach reduce the number of application migrations while maintaining low latency?
• RQ4 To what extent can programmable switches be used to incrementally upgrade existing LTE infrastructure for MEC support?
• RQ5 Can a virtualized MEC cluster abstraction improve load balancing and resource utilization in mobile edge environments?

Key findings

• The two-stage traffic steering mechanism significantly reduces the number of required application migrations by treating multiple physical MEC clusters as a single virtual cluster per region.
• MEGW enables transparent mobility support for MEC applications without modifying the LTE protocol stack, user equipment, or edge applications.
• The solution is fully compatible with existing LTE control and data planes, as verified in a testbed using OpenAirInterface’s reference 5G protocol stack.
• MEGW reduces back-haul pressure by offloading content delivery to the radio edge, improving latency and network efficiency.
• The virtual MEC cluster abstraction enhances load balancing and resource utilization, making the system more resilient to user mobility patterns.
• The implementation using P4 on programmable switches demonstrates feasibility and backward compatibility, enabling incremental deployment in existing networks.