Da Feng

Joint provisioning of lightpaths and storage in store-and-transfer wavelength-division multiplexing networks

So-called store-and-transfer wavelengthdivision multiplexing networks (STWNs) can store data and provision lightpaths at an optimal time when the wavelengths are clear of conflicts. If the wavelength utilization of the wavelength-division multiplexing network is not uniform, the embedded storage nodes can store data and the lightpath is segmented so conflicts along separate segments of the lightpath can be resolved independently. Consequently, blocked requests can be reduced, and the resource utilization of the network can be improved. In this work, we investigate three extensions of the sliding scheduled traffic model, which is a variant of advance reservations. First, we segment a lightpath with embedded storage nodes, split the transfer window into smaller ones, and sequentially assign a smaller window to each segment of the lightpath. Each segment of the lightpath has fewer hops, and thus, overall blocking can be decreased. Second, we allocate fixed time slots in continuous frames over multiple wavelengths for a time-division multiplexing (TDM) mode large data transfer. The resource allocations are more flexible in resolving conflicts, and thus, blocking can be decreased. Third, we use segmented lightpaths and TDM mode transfer along each segment of the lightpath. Since the first two extensions are orthogonal, the combination can decrease the blocking further. We propose four heuristic algorithms to evaluate the extensions. We also propose two parallel implementations of the four heuristic algorithms. The simulations show the following: the provisioning algorithms for the first two extensions have similar performances, and both decrease blocking by 20%. If the service time is long, the second one performs better. The provisioning algorithm for the last extension decreases blocking by 40%. The first extension requires 10% more storage, and the second extension requires 20% less storage. Under a large load and with 8 CPUs, the parallel implementations improve the speed by three times. It also shows that the number of time slots per frame influences the performance; the influence is not monotonic; and, if the value increases too much, the blocking rate increases by 20% and the storage size increases by 30%.

Dimensioning of the store-and-transfer WDM network with limited node storage under the sliding scheduled traffic model

The so-called store-and-transfer WDM network (STWN) can store data in source storage and provision lightpaths at an optimal time when wavelengths are clear of conflicts. Consequently, blocking of requests can be reduced and resource utilization of the network can be improved. In this work, we investigate dimensioning of STWN and propose a two-step method to jointly determine the number of wavelengths and storage size required to satisfy the demand, which is given as a load matrix with a deadline, blocking rate, and wavelength utilization. The method models the STWN as a TDM network and first obtains the required storage size by maximizing the number of fixed time slots in each period, then calculates the required number of wavelengths with the number of time slots. Numerical results show the following: high load between far apart source and destination nodes has significant impact on the required wavelengths and storage. For instance, in a 24-node topology, 20% more wavelengths and storage may be required to satisfy a biased load matrix than a randomly generated one. When calculating the required number of wavelengths, ourmethod outperforms traditional routing and wavelength assignment (RWA), because the TDM-based model supports fractional load. In the 24-node topology, 18% more wavelengths may be required byRWA than by our method. By installing amoderate amount of storage,wavelength utilization can be effectively improved. With number of wavelengths and storage size equaling 29 and 274, in the 24-node topology, utilization can reach 0.9. We also find wavelength is equivalent with storage regarding capacity of a STWN. For instance, with number of wavelengths and storage size equaling 114 and 131, in the 24-node topology, the blocking rate is the same as with number of wavelengths and storage size equaling 29 and 270.

Performance analysis of overlapping shared segment protection in Store-and-Transfer WDM networks

In this work, we investigate overlapping shared segment protection under continuous and TDM-mode sliding scheduled traffic model and compare its performance with that of shared path protection under sliding scheduled traffic model in Store-and-Transfer WDM Networks. Simulations show the following: Under continuous sliding scheduled traffic model, shared segment protection has 12% lower blocking rate than shared path protection and if number of requests is large, shared segment protection requires 20% more storage than share path protection.

Dimensioning of Store-and-Transfer WDM Network with limited storage

For large data transfers, the Store-and-Transfer WDM Network (STWN) provides storage and lightpaths jointly. Constrained delays of the transfers restrict the number of waiting requests, thus the storage size can be limited. For dimensioning of STWN, we propose a method to determine the number of wavelengths and storage size required to satisfy the demand. The demand is given as a load matrix with a common deadline and blocking, then we express the load matrix as a base load multiplying an integer matrix consisting of mutually prime elements. The dimensioning process has two steps. 1) We model STWN as TDM network and analyse channels of the TDM system as M/G/1 subsystems. We calculate approximate performance of source-destination pairs and minimize the service rate provisioned for the demand with linear search to find the maximum allowed number of time-slots. 2) With the number of time-slots fixed, we optimize routing and wavelength and time-slot assignment to minimize the number of wavelengths required to accept the integer matrix. In step 1, the approximate performance calculated includes turnout, distribution of delay and storage size. Numerical results show how to drastically decrease the required resources.

G-TEP: A GMPLS Testing and Emulation Platform

We report G-TEP which includes a GMPLS tester of highly dynamic GMPLS networks, and an emulated GMPLS network with up to 200 nodes targeted at future nation-wide deployment of ASON in China

Dimensioning of store-and-transfer WDM networks with stratified ROADM node

Abstract The so called Store-and-Transfer WDM Network (STWN) can store data in source storage and provision lightpaths at an optimal time when wavelengths are clear of conflicts. Consequently, blocking of requests can be decreased and resource utilization of the network can be improved. If traffic can be classified into delay-sensitive packet flows and deadline-constrained large data transfers, we can use a stratified node architecture to reduce cost of STWN node by reducing port count of optical and electronic switches. In this work, we investigate dimensioning of a network of flexible grid reconfigurable optical add/drop multiplexers and compare the result with that obtain by dimensioning of STWN. We propose a two-step method to determine the amount of spectrum required to satisfy the demand of large data transfers, which is given as a load matrix with a deadline and blocking rate. We also obtain the overall utilization over all links. The method firstly uses Markov chain to model variable bandwidth allocation over a lightpath and searches for amount of required spectrum of all source-destination pairs, and secondly optimizes routing and spectrum assignment for the matrix of amount of required spectrum to minimize maximum amount of spectrum used over each link. Numerical results show the following with all numbers given as absolute values: If bandwidth allocation method Min is used, the utilization can reach 0.7 and the amount of required spectrum is 30% more than that required by STWN. If bandwidth allocation method Dyna is used, the utilization can reach 0.9 and the amount of required spectrum is similar to that required by STWN. Also, under the same load, if blocking increases from 0.001 to 0.01, the amount of required spectrum decreases and the utilization increases 0.08, which denotes over-provisioning of spectrum and allowing preemption of spectrum during hours of peak traffic can effectively decrease blocking.

Overlapping shared segment protection in store-and-transfer WDM networks under sliding scheduled traffic model

Abstract The so called Store-and-Transfer WDM Network (STWN) can store data in embedded storage nodes and provision segmented lightpaths under sliding scheduled traffic model at an optimal time when wavelengths are clear of conflicts. The embedded storage nodes can also be used to segment protection paths, but those nodes must also be protected. In this work, we extend overlapping shared segment protection to provision resources under sliding scheduled traffic model. Traditionally, shared path protection is used under sliding scheduled traffic model and with link-disjoint or node-disjoint protection paths, fibre links or optical nodes including fibre links connecting them, can be protected. We compare performance of overlapping shared segment protection with shared path protection. Simulations show the following: With link-disjoint or node-disjoint protection path, overlapping shared segment protection has 30% lower blocking rate than shared path protection. On the other hand, if number of requests is large, overlapping shared segment protection requires 25% more storage than shared path protection. But if only 6 nodes of a network of 24 nodes are willing to share storage, the performance is 25% lower, while monte carlo method shows 18 nodes are required to share storage to reach stable performance. Also, if maximum number of segments increases, the blocking rate decreases and required storage size increases. In a network with diameter of 7 hops, stable performance can be reached by using 4 as maximum number of segments.

Hybrid Radio Frequency and Free Space Optical communication for 5G backhaul

5G backhaul requires high bandwidth and hybrid Radio Frequency/ Free Space Optical (RF/FSO) Communication offers Gbps links. The weather affects availability of both mmW RF and FSO links. Below cloud ceiling, the availability of hybrid RF/FSO link is above 85.7% and it can only be used for deadline-constrained large data transfers. At 17–22km above ground, the bit error rate of FSO links reaches below 10−3 and it can be used for delay-sensitive packet transfers. In this work, first, we propose a method to determine the required number of wavelengths and storage size for large data transfers and second, we propose a distributed implementation of automatic repeat request protocol to support 10Gbps bandwidth and millisecond delay for delay-sensitive packet transfers. The simulation results show the following: With link availability of 0.85, blocking rate of 0.05 can be achieved for large data transfers, which indicates 0.05 of the cost of transferring data over 5G dmW spectrum is necessary to guarantee deadline. The distributed implementation of automatic repeat request has similar performance with selective repeat. Both theoretical and simulation results show if load exceeds 0.9 of maximum load allowed by packet error probability, delay increases about 20% drastically.

Delivering Data More Efficiently in the Big Data Era

We propose a data delivery model called SSS, which seamlessly integrates switching and storage capacity with scheduling. With SSS, data can be delivered more efficiently through the network.