pith. sign in

arxiv: 1907.10669 · v1 · pith:XTHWRYHYnew · submitted 2019-07-23 · 💻 cs.NI

An Optimization-enhanced MANO for Energy-efficient 5G Networks

Francesco Malandrino , Carla-Fabiana Chiasserini , Claudio Casetti , Giada Landi , Marco Capitani This is my paper

Pith reviewed 2026-05-24 17:17 UTC · model grok-4.3

classification 💻 cs.NI
keywords 5G networksMANOenergy efficiencyoptimizationnetwork orchestrationnode activationnetwork functionstraffic routing
0
0 comments X

The pith

An optimization module integrated into 5G MANO enables joint decisions on node activation, function placement, and traffic routing to reduce energy use.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper formulates an optimization problem that accounts for 5G nodes having both forwarding and computational capabilities. It integrates a solver for this problem into the management and orchestration framework so that decisions on which nodes to activate, which network functions to run, and how to route traffic can be made together. Tests using a real OpenStack and OpenDaylight testbed plus a large emulated network drawn from an actual mobile operator show the approach beats existing methods and reaches near-optimal energy performance. A reader would care because future networks must balance data movement and processing while controlling power costs, and embedding the optimizer lets standard 5G orchestration make better choices without major redesign.

Core claim

By formulating an optimization problem that jointly selects nodes to activate, the network functions they execute, and the routes for traffic, and by embedding the solver inside the 5G MANO framework, swift high-quality decisions become possible; validation on both a real testbed and an operator-scale emulation confirms consistent outperformance of state-of-the-art alternatives together with close matching to the optimum.

What carries the argument

The optimization problem that jointly decides node activation, network function placement, and traffic routing, integrated as a module inside the 5G MANO framework.

If this is right

  • Decisions must consider both the routing capacity and the processing capacity of each node.
  • The integrated module produces decisions that can be applied directly inside standard 5G orchestration.
  • Performance gains hold on a real OpenStack/OpenDaylight testbed.
  • Performance gains also hold on a large emulation whose topology and traffic match a real mobile operator.
  • The scheme consistently beats prior alternatives while approaching the mathematical optimum.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Operators could lower ongoing energy costs by adopting the module without changing the core 5G architecture.
  • The same joint-decision style might apply to other MANO objectives such as latency or reliability.
  • If solver speed improves further, the approach could support fully dynamic, per-second re-optimization in live networks.

Load-bearing premise

The optimization problem can be solved fast enough to support real-time or near-real-time decisions inside the 5G MANO framework without simplifications that erase the energy savings.

What would settle it

A timing measurement on the testbed or emulated network showing that the solver takes longer than the MANO decision window while the resulting energy use falls short of the reported near-optimum.

Figures

Figures reproduced from arXiv: 1907.10669 by Carla-Fabiana Chiasserini, Claudio Casetti, Francesco Malandrino, Giada Landi, Marco Capitani.

Figure 1
Figure 1. Figure 1: Logical graph for vEPC. Solid lines correspond to user traffic, dashed [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: The NFV-MANO architectural framework. Source: [17] [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Example implementation of the logical graph in Fig. 1 over a physical [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: The OptiLoop strategy. We begin by obtaining an initial feasible [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Architecture (left) and topology (right) of the real-world testbed. Fig. 5(b) also indicates the paths used in our path instantiation experiments, discussed [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Architecture (left) and topology (right) of the emulation-based topology. Mininet is used to emulate a network whose topology and traffic match those [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Real-world testbed, path instantiation experiment: evolution of the [PITH_FULL_IMAGE:figures/full_fig_p010_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Mininet experiments with real-world topology: energy savings obtained as a function of traffic (left); spare computational capabilities of the active [PITH_FULL_IMAGE:figures/full_fig_p011_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Mininet experiments with real-world topology: breakdown of energy [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Mininet experiments with real-world topology: number of deployed instances for the eNB and P/S-GW VNFs (left); average number of VNFs [PITH_FULL_IMAGE:figures/full_fig_p012_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Scaled-up network topology. As in Fig. 6, blue dots indicate B/F [PITH_FULL_IMAGE:figures/full_fig_p012_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Mininet experiments with scaled-up topology: savings obtained as a function of traffic (left); spare computational capabilities of the active topology [PITH_FULL_IMAGE:figures/full_fig_p013_12.png] view at source ↗
read the original abstract

5G network nodes, fronthaul and backhaul alike, will have both forwarding and computational capabilities. This makes energy-efficient network management more challenging, as decisions such as activating or deactivating a node impact on both the ability of the network to route traffic and the amount of processing it can perform. To this end, we formulate an optimization problem accounting for the main features of 5G nodes and the traffic they serve, allowing joint decisions about (i) the nodes to activate, (ii) the network functions they run, and (iii) the traffic routing. Our optimization module is integrated within the management and orchestration framework of 5G, thus enabling swift and high-quality decisions. We test our scheme with both a real-world testbed based on OpenStack and OpenDaylight, and a large-scale emulated network whose topology and traffic come from a real-world mobile operator, finding it to consistently outperform state-of-the art alternatives and closely match the optimum.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The manuscript formulates an optimization problem for joint decisions on node activation, network function placement, and traffic routing in 5G networks to improve energy efficiency. The optimization is integrated into the MANO framework and evaluated using a real-world testbed with OpenStack and OpenDaylight, as well as a large-scale emulation based on real operator topology and traffic, where it outperforms state-of-the-art alternatives and closely matches the optimum.

Significance. If the optimization can be solved in a timely manner to support real-time decisions, this work would be significant for enabling energy-efficient 5G network management with practical validation on testbeds and emulations using real data. The dual evaluation approach (testbed plus large-scale emulation) is a strength.

major comments (1)
  1. [Abstract] Abstract: the claim that the module enables 'swift and high-quality decisions' inside the MANO framework is load-bearing for the central contribution, yet the abstract (and reader's summary of the full text) provides no information on solver runtime, scalability with network size, or how model parameters were selected; without this the integration claim cannot be assessed.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the single major comment below and agree that the abstract would benefit from additional detail to support the integration claim.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that the module enables 'swift and high-quality decisions' inside the MANO framework is load-bearing for the central contribution, yet the abstract (and reader's summary of the full text) provides no information on solver runtime, scalability with network size, or how model parameters were selected; without this the integration claim cannot be assessed.

    Authors: We agree that the abstract does not currently include quantitative information on solver runtime or scalability, which would strengthen the claim. The full manuscript reports runtime measurements from the OpenStack/OpenDaylight testbed and scalability results on the operator-derived emulated topology (with details on how parameters were derived from real traffic traces), but these are not summarized in the abstract. We will revise the abstract to add a concise statement on observed runtimes and scalability to better support the MANO integration claim. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper formulates an optimization problem for joint node activation, NF placement and routing in 5G networks, integrates the module into the MANO framework, and reports empirical results from a real OpenStack/OpenDaylight testbed plus large-scale emulation using operator topology and traffic. No load-bearing step reduces by construction to a fitted parameter, self-definition, or self-citation chain; the optimization is presented as an independent module whose outputs are validated externally against alternatives and the optimum.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no explicit free parameters, axioms or invented entities are stated. The central claim rests on the unstated assumption that the formulated ILP or similar optimization remains tractable at network scale.

pith-pipeline@v0.9.0 · 5711 in / 1176 out tokens · 16796 ms · 2026-05-24T17:17:43.111105+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

22 extracted references · 22 canonical work pages

  1. [1]

    Coopetition between network operators and content providers in SDN/NFV core networks,

    N. Gazit, F. Malandrino, and D. Hay, “Coopetition between network operators and content providers in SDN/NFV core networks,” in IEEE INFOCOM SWFAN Workshop, 2016

    work page 2016

  2. [2]

    Near optimal placement of virtual network functions,

    R. Cohen, L. Lewin-Eytan, J. S. Naor, and D. Raz, “Near optimal placement of virtual network functions,” in IEEE INFOCOM, 2015

    work page 2015

  3. [3]

    Joint Optimization of Service Function Chaining and Resource Allocation in Network Function Virtualization,

    L. Wang, Z. Lu, X. Wen, R. Knopp, and R. Gupta, “Joint Optimization of Service Function Chaining and Resource Allocation in Network Function Virtualization,” IEEE Access, 2016

    work page 2016

  4. [4]

    Deploying chains of virtual network functions: On the relation between link and server usage,

    T.-W. Kuo, B.-H. Liou, K. C.-J. Lin, and M.-J. Tsai, “Deploying chains of virtual network functions: On the relation between link and server usage,” in IEEE INFOCOM, 2016. 13

    work page 2016

  5. [5]

    Delay-aware scheduling and resource optimization with network function virtualization,

    L. Qu, C. Assi, and K. Shaban, “Delay-aware scheduling and resource optimization with network function virtualization,” IEEE Trans. on Communications, 2016

    work page 2016

  6. [6]

    An Online Stochastic Buy-Sell Mechanism for VNF chains in the NFV market,

    X. Zhang, Z. Huang, C. Wu, Z. Li, and F. C. Lau, “An Online Stochastic Buy-Sell Mechanism for VNF chains in the NFV market,” IEEE Journal on Selected Areas in Communications , 2017

    work page 2017

  7. [7]

    Energy-Aware Placement and Scheduling of Network Traffic Flows with Deadlines on Virtual Network Functions,

    N. El Khoury, S. Ayoubi, and C. Assi, “Energy-Aware Placement and Scheduling of Network Traffic Flows with Deadlines on Virtual Network Functions,” in IEEE CloudNet, 2016

    work page 2016

  8. [8]

    SDN/NFV- based Mobile Packet Core Network Architectures: A Survey,

    V . G. Nguyen, A. Brunstrom, K. J. Grinnemo, and J. Taheri, “SDN/NFV- based Mobile Packet Core Network Architectures: A Survey,” IEEE Communications Surveys Tutorials, 2017

    work page 2017

  9. [9]

    Mobile core network virtualization: A model for combined virtual core network function placement and topology optimization,

    A. Baumgartner, V . S. Reddy, and T. Bauschert, “Mobile core network virtualization: A model for combined virtual core network function placement and topology optimization,” in IEEE NetSoft, 2015

    work page 2015

  10. [10]

    Joint Bearer Aggregation and Control- Data Plane Separation in LTE EPC for Increasing M2M Communication Capacity,

    G. Hasegawa and M. Murata, “Joint Bearer Aggregation and Control- Data Plane Separation in LTE EPC for Increasing M2M Communication Capacity,” in IEEE GLOBECOM, 2015

    work page 2015

  11. [11]

    QoS management SDN- based for LTE/EPC with QoE evaluation: IMS use case,

    S. Khairi, M. Bellafkih, and B. Raouyane, “QoS management SDN- based for LTE/EPC with QoE evaluation: IMS use case,” in SDS, 2017

    work page 2017

  12. [12]

    SDN-based vs. software-only EPC gateways: A cost analysis,

    X. An, W. Kiess, J. Varga, J. Prade, H.-J. Morper, and K. Hoffmann, “SDN-based vs. software-only EPC gateways: A cost analysis,” in IEEE NetSoft, 2016

    work page 2016

  13. [13]

    Modeling and Dimensioning of a Virtualized MME for 5G Mobile Networks,

    J. Prados-Garzon, J. J. Ramos-Munoz, P. Ameigeiras, P. Andres- Maldonado, and J. M. Lopez-Soler, “Modeling and Dimensioning of a Virtualized MME for 5G Mobile Networks,” IEEE Trans. on Veh. Tech., 2017

    work page 2017

  14. [14]

    http://www.openairinterface.org

    OpenAirInterface: 5G software alliance for democratising wireless in- novation. http://www.openairinterface.org

    work page

  15. [15]

    Network function placement on virtualized cellular cores,

    D. Dietrich, C. Papagianni, P. Papadimitriou, and J. S. Baras, “Network function placement on virtualized cellular cores,” in COMSNETS, 2017

    work page 2017

  16. [16]

    Optimization-in-the-Loop for Energy-Efficient 5G,

    F. Malandrino, C. F. Chiasserini, C. E. Casetti, and G. Landi, “Optimization-in-the-Loop for Energy-Efficient 5G,” in IEEE WoW- MoM, 2018

    work page 2018

  17. [17]

    (2017) Network Functions Virtualisation (NFV); Management and Orchestration

    ETSI. (2017) Network Functions Virtualisation (NFV); Management and Orchestration. http://www.etsi.org/deliver/etsi gs/NFV-MAN/001 099/001/01.01.01 60/gs NFV-MAN001v010101p.pdf

    work page 2017

  18. [18]

    Cvxgen: A code generator for embedded convex optimization,

    J. Mattingley and S. Boyd, “Cvxgen: A code generator for embedded convex optimization,” Optimization and Engineering , 2012

    work page 2012

  19. [19]

    Boyd and L

    S. Boyd and L. Vandenberghe, Convex optimization . Cambridge university press, 2004

    work page 2004

  20. [20]

    Deliverable D3.2: Final XFE/XCI design and specification of southbound and northbound interfaces

    5G Crosshaul Project. Deliverable D3.2: Final XFE/XCI design and specification of southbound and northbound interfaces. http://5g-crosshaul.eu/wp-content/uploads/2018/01/5G-CROSSHAUL D3.2.pdf

    work page 2018

  21. [21]

    Cisco Visual Networking Index,

    Cisco, “Cisco Visual Networking Index,” 2017

    work page 2017

  22. [22]

    It’s kind of fun to do the impos- sible with DPDK

    Lagopus Project. It’s kind of fun to do the impos- sible with DPDK. https://www.slideshare.net/lagopus/ dpdk-summit-2015-its-kind-of-fun-to-do-the-impossible-with-dpdk. 14

    work page 2015