Published on Feb 12, 2016
In this seminar, we give an introduction to optical burst switching (OBS) and compare it with other existing optical switching paradigms. Basic burst assembly algorithms and their effect on assembled burst traffic characteristics are described first. Then a brief review of the early work on burst transmission is provided followed by the description of a prevailing protocol for OBS networks called Just-Enough-Time (JET).
Description of Optical Burst Switching
With recent advances in wavelength division multiplexing (WDM) technology, the amount of raw bandwidth available in fiber links has increased by many orders of magnitude. Meanwhile, the rapid growth of Internet traffic requires high transmission rates beyond a conventional electronic router's capability. Harnessing the huge bandwidth in optical fiber cost-effectively is essential for the development of the next generation optical Internet.
Several approaches have been proposed to take advantage of optical communications and in particular optical switch¬ing. One such approach is optical circuit switching based on wavelength (A) routing whereby a lightpath needs to be established using a dedicated wavelength on each link from source to destination. Once the connection is set up, data remains in the optical domain throughout the lightpath. An alternative to optical circuit switching is optical packet switching. In optical packet switching, while the packet header is being processed either ail-optically or electronically after an Optical/Electronic (O/E) conversion at each intermediate node, the data payload must wait in the fiber delay lines and be forwarded later to the next node.
Usually, assembly algorithms can be classified as timer-based, burstlength-based and mixed timer/burst length-based ones.
In the timer-based scheme, a timer starts at the beginning of each new assembly cycle. After a fixed time T, all the packets that arrived in this period are assembled into a burst. In the burstlength-based scheme, there is a threshold on the (minimum) burst length. A burst is assembled when a new packet arrives making the total length of current buffered packets exceed the threshold.
The time out value for timer-based schemes should be set carefully. If the value is too large, the packet delay at the edge might be intolerable. If the value is too small, too many small bursts will be generated resulting in a higher control overhead. While timer-based schemes might result in undesirable burst lengths, burstlength-based assembly algorithms do not provide any guarantee on the assembly delay that packets will experience. To address the deficiency associated with each type of the assembly algorithms mentioned above, mixed timer/threshold-based assembly algorithms were proposed in. For example, in, a burst can be sent out when either the burst length exceeds the desirable threshold or the timer expires.
Burst Reservation Protocols
It is evaluated two burst level admission control mechanisms for ATM networks: tell-and-wait and tell-and-go. In the former, when a source has a burst to transmit, it first tries to reserve the bandwidth/wavelength from the source to its destination by sending a short 'request' message. Every intermediate node receiving this message will make a reservation on a specific output link. If the requested bandwidth is successfully reserved on all the links along the path, an ACK will be sent back to inform the source to send out the burst immediately; Otherwise, a NAK will be returned to release the previously reserved bandwidth, and initiate the retransmission of the 'request' message after a backoff time.
In tell-and-go, on the other hand, the source transmits bursts without making any bandwidth reservation in advance. At an intermediate node, the burst needs to be delayed before the switch control unit makes an appropriate reservation on an outgoing link. If the reservation fails at any intermediate node, a NAK will be sent back to the source to initiate the retransmission of the burst after a backoff time.