opex savings by reduction of stock of spare parts with
play

OpEx savings by reduction of stock of spare parts with Sliceable - PowerPoint PPT Presentation

OpEx savings by reduction of stock of spare parts with Sliceable Bandwidth Variable Transponders. Beatriz de la Cruz, Oscar Gonzlez de Dios, Victor Lopez, Juan Pedro Fernndez-Palacios Index Overview 01 SBVTs 02 Case Study 03 Results


  1. OpEx savings by reduction of stock of spare parts with Sliceable Bandwidth Variable Transponders. Beatriz de la Cruz, Oscar González de Dios, Victor Lopez, Juan Pedro Fernández-Palacios

  2. Index Overview 01 SBVTs 02 Case Study 03 Results 04 05 Conclusions Core Networks Evolution 2 Telefónica I+D

  3. 01 Overview

  4. Overview Stock Management Rationale The importance of having the suitable equipment at the right time. Reducing operational expenditures by an optimal stock study, based on Objective centralized model. • Minimum stock number required in order to keep the service. • Stock number for each kind of transponder Results • Percentage of time which a certain stock number keeps the service Expected Reducing expenditures related network maintenance and reparation increasing thus indirectly the benefits. benefits Core Networks Evolution 4 Telefónica I+D

  5. Stock of spare parts with SBVT problem Problem definition. • Emerging technologies can help to deal with the ever-increasing demand: § Elastic Optical Network (EON). § Sliceable Bandwidth Variable Transponder (SBVT). • Different studies conclude SBVT allows a reduction in CAPEX. • Our work: quantify the reduction of network maintenance and reparation related OpEx by using SBVT. § Focused on cost related to keeping a stock of spare parts. • Stock of spare parts for replacement needs to be maintained in case of a failure in a network element. • Centralized stock model. • Analyze how equipping a network with Sliceable Bandwidth Variable Transponders instead of fixed rate transponders of multiple rates reduces the maintenance cost. Core Networks Evolution 5 Telefónica I+D

  6. 02 SBVTs

  7. Current node structure without SBVTs • Node model for the study: Each line card needs a dedicated FW card of 40 or 100G switching capacity FW Card FW Card IP/MPLS Router Line Cards Chassis (including FW card) · · · IP/MPLS Tx Rx Tx Rx Transceiver SR Tx Rx Tx Rx Termninals Transceiver SR Tx Tx Rx Rx · · · WDM Transponders WDM Transponder Tx Rx Tx Rx A different WDM transponder per destination Core Networks Evolution 7 Telefónica I+D

  8. Node structure with SBVTs • Node models for the study: FW card with higher switching capacity (i.e. 400Gbps, 1Tbps) IP/MPLS Router Line Cards Chassis FW Card IP/MPLS (including FW card) Tx / Rx Tx / Rx Ethernet Switch or OTN OTN framing Crossconnect added Limited Number of destinations S-BVT Transponder Tx/Rx Multiple Destinations Core Networks Evolution 8 Telefónica I+D

  9. 03 Case Study

  10. Case Study • Spanish National Network: TR13 Scenario 1: Full Mesh Topology. § 2 12 Scenario 2: IP Top. Shortest path TR14 § 1 TR11 3 11 13 4 TR3 TR4 • IP nodes: TR1-TR14. 16 14 TR12 15 10 6 TR2 5 9 TR6 TR1 18 • Optical nodes: 1-30 8 17 7 20 TR5 Add/drop Node 25 26 19 • Each IP node is attached to an optical 27 OXC 21 TR10 22 node. TR7 28 TR9 23 29 • Studied cases: TR8 24 30 Case 1: Fixed traditional transponders. § Case 2: SBVTs. § Core Networks Evolution 10 Telefónica I+D

  11. Procedure Generate random failures in time, based on Failures MTBF and following an exponential distribution. Get replacement information, knowing the necessary Replace ments time to replace one transponder. Stock counter: +1 if a fail happens, -1 if a replacement Stock happens. Obtain peak value à minimum stock number Results Achieve the percentage of time which a certain stock number keeps the service, knowing the peak value of minimum stock. Core Networks Evolution 11 Telefónica I+D

  12. Input Data What is the required information? • Transponders: § Number of transponders § Fixed grid transponders: § Sliceable Bandwidth Variable Transponders § Cost Cost 40Gb/s, 2500km, 50 GHz 6 100Gb/s, 2000km, 50 GHz 15 • Time: 400Gb/s, 75GHz, 500km 22 § Reposition time à 3 months § Operational time à 10 years • Failures: § Mean Time Between Failures à 5 years Core Networks Evolution 12 Telefónica I+D

  13. 04 Results

  14. Number of spare transponders Percentage of failures that can be repaired maintaining a certain number of spare transponders in a Full Mesh scenario for year 2014. Non-sliceable transponders SBVTs 100,5 100,5 100,0 100,0 99,5 99,5 40 Gbps (%) Availability 100 Gbps 99,0 99,0 99,999 % 400 Gbps (%) Availability (%) Availability 98,5 98,5 99,999 % 98,0 98,0 97,5 97,5 97,0 97,0 96,5 96,5 96,0 96,0 0 1 2 3 4 5 6 7 8 9 10 11 12 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Number of Transponders Number of Transponders 30 Fixed Vs 9 SBVTs Lower number of spare transponders are necessary in the case of using SBVT to keep 99.999% availability. Core Networks Evolution 14 Telefónica I+D

  15. Service availability from 2014 to 2020 Number of Stock required to keep 99.999% of availability v Full Mesh v IP Topology In both cases (Full mesh and IP) along the years higher number of total fixed, transponders are needed versus SBVT to keep 99.999% of availability. Core Networks Evolution 15 Telefónica I+D

  16. SBVT Target Cost Full Mesh and IP Topology Target cost of SBVT to save 30% operational expenditures. • Possible cost of SBVT can vary between 22 and 30. 22 is the cost of 400Gbps fixed transponders § 30 imply an increase to 36% in the cost of the non sliceable § transponder. • The peak target cost value is reached in 2015 (12.5 Tbps) Core Networks Evolution 16 Telefónica I+D

  17. 05 Conclusions

  18. Conclusions Summary of obtained results. • Operational Expenditures can be reduced base on the previous results obtained: § Lower number of spare transponders are necessary in the case of using SBVT to keep 99.999% availability § In both cases (Full mesh and IP) along the years higher number of total fixed keep being necessary versus SBVT. § Possible cost of SBVT can vary between 22 and 30 which implies an increase to 36% in the cost of the fixed transponder. Core Networks Evolution 18 Telefónica I+D

  19. Core Networks Evolution 19 Telefónica I+D

Recommend


More recommend