Name
Abstract | ||
|---|---|---|
This paper presents a new weight incidence representation of Dynamic wavelength addressing in optical fiber networks utilizing wavelength division multiplexing (WDM) can form the basis for a high-performance, high-bandwidth, low-latency any-to-any interconnection network. WDM optical fiber networks exploit the fact that photons of different wavelengths do not normally interact, thereby enabling the transmission of many channels of data in parallel in the same fiber by using photons of a different wavelength for each channel. Wavelength addressing is a networking concept that utilizes WDM technology to enable direct routing of data. Each node in the network is assigned a specific wavelength that is considered its address. Data transmitted in a certain wavelength on the network is read only by the node that was assigned that wavelength as its address. We demonstrate the effectiveness of this approach by considering the interconnection needs of a storage area network (SAN). A SAN is a special-purpose interconnection scheme that enables computer systems to view remote storage devices as local The underlying principle of a dynamic wavelength addressing SAN (WASAN) is to add the existing SAN architecture a dynamic multi-fiber WDM loop with dynamic add/drop capability to each node. A client requesting a data package sends its request to the SAN server to which it is connected. The SAN management then allocates a fiber in the loop, and a wavelength on this fiber to this request. This fiber-wavelength pair now constitutes a circuit that connects the client to the storage device that contains the requested data package. Data can now flow directly and continuously from the storage device to the client. Upon completion of the operation, the fiber-wavelength pair is released to the SAN management to be used for a different task. The effectiveness of the WASAN is investigated in a series of simulations Indeed the WASAN performance is superior to the respective SAN of similar sizes and with equal work loads. It is clearly seen that the WASAN performance is significantly less sensitive when the average package size is increased. Furthermore, adding load to the WASAN architecture in the form of extra transmitting nodes does not create the congestion that is formed in the regular SAN architecture, therefore yielding better performance and greater scalability in system size as well as data size. A straightforward implementation of WASAN would equip each of the nodes in the network with an expensive tunable laser source so that any wavelength can be assigned arbitrarily to each node according to availability by the WASAN management. We investigate an alternative scheme: the laser power grid. The laser power grid is a fiber network adjunct to the circuit switching network that carries raw optical power at the wavelengths that are used by the circuit switching network. The raw optical power is supplied from a battery of single wavelength lasers. Each node is equipped with a special wavelength allocation device that upon request transfers optical power at a prescribed wavelength to the node transmitter. This obviates the need for tunable laser sources at each node, which significantly boosts the cost effectiveness of the WASAN. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1007/978-3-642-10442-8_18 | OSC |
Keywords | Field | DocType |
certain wavelength,storage area networks,san management,dynamic optical circuit switching,dynamic wavelength,laser power grid,single wavelength laser,prescribed wavelength,different wavelength,raw optical power,wasan performance,storage device,low latency,storage area network,optical fiber,tunable laser,cost effectiveness,circuit switched | Wavelength-division multiplexing,Optical fiber,Circuit switching,Fiber in the loop,Tunable laser,Electronic engineering,Engineering,Interconnection,Storage area network,Scalability | Conference |
Volume | ISSN | Citations |
5882 | 0302-9743 | 0 |
PageRank | References | Authors |
0.34 | 1 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Aharon J. Agranat | 1 | 0 | 0.68 |
| Noam Sapiens | 2 | 0 | 0.34 |
| Larry Rudolph | 3 | 37 | 5.01 |