Packet mode scheduling in buffered crossbar (CICQ) switches

Abstract

Abstract-Buffered crossbars can directly switch variable size packets but they require significant crosspoint buffering to do so, especially when the traffic includes large packets. When we cannot afford large crosspoint buffers we are forced to restrict the maximum internal transfer unit by segmenting packets. Packet segmentation implies a reassembly delay cost which is an issue in systems requiring low latency. We drastically reduce reassembly delay by applying packet mode scheduling to the buffered crossbar architecture. Packet mode scheduling has been studied in input queued switches: when the central switch scheduler establishes a connection from a switch input to a switch output port, it maintains that connection until all the cells of the packet are switched. In buffered crossbars the scheduling is distributed at switch input and output ports, thus the extension is not trivial. We synchronize the input and output port schedulers so as whenever their independent decisions result to an input-output port pairing they maintain that pairing for the lifetime of the packet transmission. Using simulation we study our system's performance. We show that reassembly delay is significantly reduced, especially under light loads.

Packet-Mode Asynchronous Scheduling Algorithm for Partially Buffered Crossbar Switches

“…Since the length of IP packets typically varies from 40 to 1,500 bytes [5], we assume that the average packet length is 770 bytes, which is equal to 1/p. Using the result in Theorem 2, we know that the conflict-free probability is higher than 99%.…”

“…Because the geometric packet length distribution used in the analysis of Section III-B may have an arbitrary long packet length, it is not practical. We set the packet length to be distributed between 40 and 1,500 bytes [5]. Since in a real network, packet arrival is typically bursty and packets are highly correlated, we use a two-state Markov Modulated Poisson Process (MMPP) to model the incoming traffic [2].…”

“…The reason is that an input cannot sustain simultaneous transmission to two outputs. The restriction can be implemented by using a lock similar to that in [5]. If an input is sending a packet to a crosspoint buffer, which is sending the same packet to the output using cutthrough switching, the input is locked.…”

“…Therefore, a connection between an input-output pair has to be maintained until all cells of a packet finish transmission. Packet-mode scheduling has unique advantages over cell-mode scheduling [2], [5], such as high throughput, low delay, and reduced hardware cost. However, synchronized packet-mode scheduling of non-buffered crossbar switches has some drawbacks.…”

Asynchronous Operation of Bufferless Crossbars

“…Our study was motivated by our previous work on variablesize packet switching in buffered crossbars [9], [10], [11]. Crosspoint buffering allows for temporarily conflicting inputoutput matchings and, this, in turn, renders asynchronous operation straightforward.…”