Meeting the requirements to deploy cloud RAN over optical networks - elastic optical network resources meet compute resources
1 like98 views
The document discusses deploying cloud RAN over optical networks, focusing on the requirements for dynamic and elastic bandwidth. It proposes using sliceable bandwidth-variable transponders to optimize costs and capacity for distributed radio heads (RRHs) and highlights strategies for minimizing capital and operational expenses in the network configuration. Key insights include the cost implications of different transponder types and the energy-saving potential by consolidating operations at central offices.
Meeting the requirements to deploy cloud RAN over optical networks - elastic optical network resources meet compute resources
- 1. Meeting the requirements to deploy cloud RAN over optical networks: elastic optical network resources meet compute resourcesSabidur Rahman Netlab, UC Davis [email protected] L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 2. Introduction • RRHs can share virtualized BBU pools • Dynamicity, fine granularity, and elasticity • Proposed Sliceable Bandwidth-Variable Transponders • Which COs to place the SBVT to minimize CAPEX and OPEX • SBVT vs. Fixed Transponders L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 3. Motivation C-RAN requires dynamic, elastic, fine granularity from 10Gb/s to 100 Gbps L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 4. Types of Transponders To interface the optical layer, there are several types of transponders that can be used in both, front and backhaul: • Fixed Transponders (FT) that transmit at a fixed bitrate, e.g., 40Gb/s, • Bandwidth-Variable Transponders (BVT) that can adapt its bitrate up to a maximum capacity e.g., 400Gb/s, and • Sliceable Bandwidth-Variable Transponders (SBVT) that can be shared among a number of optical connections.
- 5. SBVT • Multiple bit rates (10 Gbps to 1 Tbps) • Dynamically changeable modulation formats • Sub-channels and super channels allows finer granularity: can avoid the need for grooming • Redo the DSP of some slices without impacting others • Dynamic Optical Arbitrary Waveform Generation and Measurement (DOAWG/DOAWM) Number of slices: 4->3->5 L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 6. Proposed SBVT Architecture Electric core: generation of sub- and super- channels NC&M: how many subchannels? Which modulation format? Select two tones from the OFC using first WSS Two Modulators forms two phase-coherent spectral slices Second WSS combines two slices into large-bandwidth superchannel L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 7. CRAN CAPEX minimization problem Given: • A set of geographically distributed RRHs H • The tuple <αh, βh, γh> representing the required capacity by RRH h for CPRI, S1, and X2 interfaces respectively, pre-computed in advance. • A set V of VMs’ configurations with capabilities for BBU pools virtualization; each VM configuration v is defined by its cost κv and its number of BBUs it can virtualize λv; let us assume that one BBU can serve one RRH • A set of transponders P; each transponder p consists of a set of DSP modules D(p), where the capacity of each module is φp and its cost κp; since gray or colored transponders may be considered to support the different interfaces, the parameters δpCPRI, δpS1, δpX2 indicate if p can support CPRI, S1 or X2 interface links respectively. • A set of line cards C; each line card c can support one type of transponder, and it is defined by its cost κc and the number of ports to plug-in transponders ξcp. • A set of MPLS equipment E • A set O with main COs; each main CO can be equipped with a predefined configuration of VMs and with a MPLS switch. • O(h) represents the subset of main COs that can be reached by RRH h without exceeding delay imposed by CPRI requirements. • U(o) accounts for the subset of main COs that can be reached from main CO o without violating X2 delay constraints. Neighbor COs • A core CO with functions for MME, S-GW, along with others Output: the VMs’ configurations and MPLS equipment, lines cards and transponders to install in each main CO. Objective: minimize the cost of VMs’ configurations, MPLS equipment, line cards and transponders used L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 8. Results(1) L. Velasco, A. Castro, A. Asensio, M. Ruiz, G. Liu, C. Qin, R. Proietti, and S. J. B. Yoo, "Meeting the requirements to deploy cloud RAN over optical networks." Journal of Optical Communications and Networking, vol. 9, no. 3, pp. 22-32, Feb. 2017.
- 9. Results(2) Take away: • FTs becomes more costly at higher load • Higher SVBT capacity saves more cost Take away: • Smaller MPLS switches saves cost
- 10. Results(3) Take away: Shutting down COs saves power (a) Fully centralized (2CO) lowest number of transponders (b) 2CO has high power consumption from large switch
- 11. Next steps • Study and evolve to define the problem statement • Sabidur Rahman • Netlab, UC Davis • [email protected] Thanks!