In this paper we propose a DC network (DCN) architecture that interconnects servers in the intra-rack and inter-rack domain, utilizing optical switching at each domain. The proposed interconnection techniques are studied as an intermediate step before migrating the entire DCN to all-optical schemes. Unlike other studies, we study the server-to-server communication across the whole DCN. For the performance evaluation we produce numerical results for throughput and end-to-end delay for three traffic classes co-existing in DCN s. The numerical analysis reveals that bandwidth utilization reaches 90% and 100% in the intra- and inter- domain respectively. Meanwhile, the maximum end-to-end delay for the highest priority packets under congested load is lower than 0.56 and 0.41 µs for the two examined intra-rack capacity scenarios of 400 and 600 Gbps respectively. A comparative study shows that our solution can effectively interconnect up to 10000 servers with lower environmental footprint and end-to-end delay than other DCN s.

The packet rejection at destination under asynchronous transmission in high-speed optical fiber networks is examined in this study. Especially, we propose an asynchronous transmission protocol and a station network interface suitable for local area networks of passive star topology aiming to reduce the rejection probability at destination due to the receiver collisions. We assume that each station uses more than one tunable receivers to be able to receive more than one packets per time instant. In this way, we provide essential performance improvement by means of rejection probability reduction and throughput enhancement. The adopted network configuration uses a separate control wavelength to coordinate the data packets transmissions. An analytical framework is adopted to quantify the receiver collisions effect on the protocol performance considering Poisson arrivals and finite station population. Numerical results are comparatively examined with the single tunable receiver per station case, for variable number of data channels and stations.

In this study, we explore the performance optimization conditions of a WDM network of passive star topology that uses a Multi-channel Control Architecture (MCA) to avoid both the data channels and destination conflicts. The adopted synchronous WDMA scheme exploits the propagation delay to assign the data channels to the stations whose control packets transmission was successful. Thus, effective bandwidth utilization is achieved. An approximate analysis based on Poisson statistics is developed in order to explore the performance measures optimization.

The key parameters affecting the behavior of a passive star WDM network are: control channel collisions, data channel collisions, propagation delay latency and destination conflicts. In this paper, we try to exploit the propagation delay latency as an acknowledgment time period to introduce a new network architecture and a suitable access protocol in order to avoid data channel collisions, achieving performance measures improvement and enhancing in this way bandwidth utilization.

The key idea of this study is the adoption of an asymmetric access strategy suitable for a WDM network architecture of passive star topology that uses the Multichannel Control Architecture (MCA). The asymmetric access strategy determines that the MCA exploits a separate control channel for the transmission of free stations, while the remaining control channels are used by the backlogged stations. The exchanged control information over the MCA is utilized to properly assign the data channels among the stations for transmission, improving data channels utilization. In addition, the access scheme effectively exploits the propagation delay latency as an appropriate attribute in order to totally avoid both the data channels collisions and the destination conflicts. Thus, we achieve essential performance improvement and bandwidth utilization enhancement. The proposed access strategy belongs to the synchronous transmission WDMA protocols. An analytic Markovian model for finite population without approximations is developed to derive the performance measures. The effect of receiver collisions is analytically examined. Extensive comparative results are investigated for different network parameters: stations population and MCA channels.

In this paper, an efficient traffic allocation technique and a slotted delay-sensitive WDMA protocol suitable for ring metropolitan area networks is proposed. Our purpose is to improve the limited bandwidth utilization that many WDMA protocols for MANs introduce, especially at high loads. This is achieved by: 1) applying an access algorithm to avoid data wavelengths and receiver collisions, and 2) introducing a multiple buffer architecture at each node with an effective buffer selection for transmission scheme that combines the priority criteria of receiver collisions avoidance and packet age. Thus, we obtain not only dropping probability and delay reduction, but also significant throughput enhancement. The required number of buffers per node to maximize throughput is investigated. Performance evaluation is provided by a discrete event simulation model based on self-similar traffic. Comparative results prove that the proposed architecture manages efficient bandwidth utilization as compared with other WDMA protocols, especially under high loads.