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

OpEx savings by reduction

• f stock of spare parts with

Sliceable Bandwidth Variable Transponders.

Beatriz de la Cruz, Oscar González de Dios, Victor Lopez, Juan Pedro Fernández-Palacios

2 Core Networks Evolution Telefónica I+D

01 02 03 04

Index

Overview SBVTs Case Study Results 05 Conclusions

01

Overview

4 Core Networks Evolution Telefónica I+D

Overview

Stock Management

Rationale

The importance of having the suitable equipment at the right time.

Objective Results

Reducing operational expenditures by an optimal stock study, based on centralized model.

• Minimum stock number required in order to keep the service.
• Stock number for each kind of transponder
• Percentage of time which a certain stock number keeps the service

Reducing expenditures related network maintenance and reparation increasing thus indirectly the benefits.

Expected benefits

5 Core Networks Evolution Telefónica I+D

Stock of spare parts with SBVT problem

• 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.

Problem definition.

02

SBVTs

7 Core Networks Evolution Telefónica I+D

Current node structure without SBVTs

• Node model for the study:

Tx Rx

Transponders IP/MPLS Chassis WDM Termninals

Tx

Rx Tx Rx

IP/MPLS Router Line Cards (including FW card) Transceiver SR Transceiver SR WDM Transponder

Tx Rx Tx Rx

Tx

Rx Tx Rx Tx Rx

Each line card needs a dedicated FW card

• f 40 or 100G switching capacity

· · · · · ·

FW Card FW Card

A different WDM transponder per destination

8 Core Networks Evolution Telefónica I+D

Node structure with SBVTs

• Node models for the study:

IP/MPLS Chassis IP/MPLS Router Line Cards (including FW card) S-BVT Transponder

Tx/Rx

FW card with higher switching capacity (i.e. 400Gbps, 1Tbps) Multiple Destinations

Ethernet Switch

• r OTN

Crossconnect

Limited Number of destinations

Tx / Rx FW Card Tx / Rx

OTN framing added

03

Case Study

10 Core Networks Evolution Telefónica I+D

Case Study

• Spanish National Network:

§ Scenario 1: Full Mesh Topology. § Scenario 2: IP Top. Shortest path

• IP nodes: TR1-TR14.
• Optical nodes: 1-30
• Each IP node is attached to an optical

node.

• Studied cases:

§ Case 1: Fixed traditional transponders. § Case 2: SBVTs.

1 2 3 4 12 5 6 10 11 8 7 9 18 25 16 13 14 15 17 19 20 21 22 26 27 28 23 29 24 30 TR11 TR12 TR13 TR14 TR3 TR4 TR5 TR6 TR1 TR2 TR7 TR8 TR9 TR10 Add/drop Node OXC

11 Core Networks Evolution Telefónica I+D

Procedure

Failures Replace ments Stock

Results

Generate random failures in time, based on MTBF and following an exponential distribution. Get replacement information, knowing the necessary time to replace one transponder. Stock counter: +1 if a fail happens, -1 if a replacement

• happens. Obtain peak value à minimum stock number

Achieve the percentage of time which a certain stock number keeps the service, knowing the peak value of minimum stock.

12 Core Networks Evolution Telefónica I+D

Input Data

• Transponders:

§ Number of transponders

§ Fixed grid transponders: § Sliceable Bandwidth Variable Transponders

§ Cost

• Time:

§ Reposition time à 3 months § Operational time à 10 years

• Failures:

§ Mean Time Between Failures à 5 years

What is the required information?

Cost 40Gb/s, 2500km, 50 GHz 6 100Gb/s, 2000km, 50 GHz 15 400Gb/s, 75GHz, 500km 22

04

Results

14 Core Networks Evolution Telefónica I+D

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.

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 96,0 96,5 97,0 97,5 98,0 98,5 99,0 99,5 100,0 100,5

(%) Availability Number of Transponders 40 Gbps 100 Gbps 400 Gbps 99,999 %

1 2 3 4 5 6 7 8 9 10 11 12 96,0 96,5 97,0 97,5 98,0 98,5 99,0 99,5 100,0 100,5

(%) Availability Number of Transponders (%) Availability 99,999 %

Non-sliceable transponders SBVTs

Lower number of spare transponders are necessary in the case of using SBVT to keep 99.999% availability. 30 Fixed 9 SBVTs

Vs

15 Core Networks Evolution Telefónica I+D

Service availability from 2014 to 2020

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.

Number of Stock required to keep 99.999% of availability

v Full Mesh v IP Topology

16 Core Networks Evolution Telefónica I+D

SBVT Target Cost

• 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)

Full Mesh and IP Topology

Target cost of SBVT to save 30% operational expenditures.

05

Conclusions

18 Core Networks Evolution Telefónica I+D

Conclusions

• Operational Expenditures can be reduced base on the previous results
• btained:

§ 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.

Summary of obtained results.

19 Core Networks Evolution Telefónica I+D