*4.2. Evaluation of the B-REP Mechanism*

The B-REP algorithm implementation uses the SPF algorithm in conjunction with any type of LS routing protocol, although we used OSPF for the pilot implementation. The B-REP SPF algorithm is applied to calculate the alternative shortest path used in the event of a failure. The main advantage of the B-REP FRR mechanism is that the algorithm implements an efficient and standardized way to mark an alternative path using the B-S field.

In the B-REP we use the bit-string to exactly define hop-by-hop routing behavior, where due to B-S we can precisely define the whole alternative path of routers chain. This feature might be used for an administrator to manually configure the alternative route in the event of a need.


**Table 6.** Output from the OMNeT++ simulation.

Possibility of the explicitly defined alternative path can not only define the backup path close to a failed element in the network but also across the whole area which statistically can also be damaged.

Compared to other existing FRR mechanisms, the B-REP mechanism brings the new approach of defining alternative path (bit-string) into the FRR area and provides advantages in comparison with other mechanisms such as custom alternative path, easy implementation into existing architecture because of bit-string and 100% repair coverage. A summary of the advantages and disadvantages of the B-REP mechanism is given in Table 7.


**Table 7.** Properties of the B-REP algorithm.

A more exact comparison of the selected features with other existing FRR solutions is presented in Table 8 below. In this table we compare possibility of repairing all failures within the network, precomputing, modification of packets and support of custom alternative path.


**Table 8.** Comparison of the B-REP mechanism with existing solutions.

The biggest time-consuming operation during the FRR process is the detection of link or node failure. For these purposes, the bidirectional forwarding detection (BFD) protocol is used. The BFD protocol is standardized by IETF in RFC 5880. Usually, the detection of the failure by BFD protocol is less than 30 ms depending on the timer settings. Another part of rerouting is switching to an alternate FRR path.

Existing proactive FRR solutions calculate alternative path in advance. Therefore, the rerouting time of the specific FRR mechanism is minimal because the alternative path is prepared and switchovers to that path immediately. B-REP algorithm calculates alternative path in advance, therefore its speed of recovery after link or node failure depends only on the time of failure detection, as is characteristic of proactive mechanisms.

#### **5. Conclusions**

The paper presents the bit-repair (B-REP) FRR mechanism, which provides advanced fast reroute solutions for IoT and IP network infrastructures. B-REP uses a standardized BIER header with a special bit-string field. That allows us to use a standardized header and its fields to define an alternative path as well as to transfer user data. The bit-string, in addition, allows us to efficiently define an exact alternative FRR path, which can be calculated by the Dijkstra algorithm or even manually defined by the administrator.

Some existing mechanisms, such as LFA or remote LFA calculate an alternative path according to specific metric conditions. However, in topologies with inappropriate metrics, these mechanisms are not able to choose an alternative path, but our algorithm is. We also add the ability to ignore metric-based calculations in our proposal, allowing us to select any possible physical alternative path. This mechanism can provide link or node protection and is suitable for any link-state protocols. These properties of the B-REP algorithm make it possible to achieve full repair coverage, which provides the protection against all possible failures in the network if a physically alternative path is presented.

The B-REP mechanism, as a proactive FRR mechanism, shares with other FRR mechanisms the properties of this family, which are identified as limitations. This includes CPU consumption during preliminary calculations and memory consumption for storing them. With FRR mechanisms that create a remote alternate path, B-REP uses the encapsulation of the original packets, which of course increases the overhead of the transmitted data. However, this limitation is partially addressed by using the Bit-String to define the entire transmission path. The speed of existing proactive FRR mechanisms including B-REP is similar, but the differences are in the way how they calculate alternative FRR path, how effective the results are and how the alternative path is constructed. According to our research, there is no FRR mechanism of all solutions. All of them have some advantages and disadvantages and B-REP brings his perspective on the solution to the issue.

The B-REP mechanism was fully implemented, and its correctness was tested using the OMNeT++ discrete event simulator. We have performed various extensive tests of the implementation in different network topologies, which validated the functional correctness of all B-REP sides. The use of Bit-String is unique, and it is possible to apply it in WSN networks, IoT design, and other areas as well, which will be studied in future work. Besides that, our future work will focus on the investigation of other related B-REP research issues, such as the addressing of multiple error occurrence, B-REP resource demands, and the B-REP bit-string-based source routing. We are also preparing the implementation of several existing FRR mechanisms into the OMNeT++ simulation tool, which will allow a better comparison of existing solutions and bring new knowledge in the field.

**Author Contributions:** Conceptualization, J.P. and P.S.; software, J.P.; validation, J.P. and P.S.; formal analysis, J.P., P.S., O.Y., I.B., M.H.; investigation, J.P.; resources, J.P.; data curation, J.P.; writing—original draft preparation, J.P., P.S., O.Y.; writing—review and editing, J.P., P.S., O.Y., I.B., M.H.; investigation, J.P.; visualization, J.P. and P.S.; supervision, J.P. and P.S.; project administration, J.P. and P.S.; funding acquisition, J.P., P.S. and I.B. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Acknowledgments:** This publication has been published with the support of the Operational Program Integrated Infrastructure within project: "Výskum v sieti SANET a možnosti jej d'alšieho využitia a rozvoja/ Research in the SANET network and possibilities of its further use and development", ITMS code 313011W988, co-financed by the ERDF.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
