Matching output queueing with a combined input/output-queued switch
read more
Citations
The iSLIP scheduling algorithm for input-queued switches
Algorithmics of Matching Under Preferences
The throughput of data switches with and without speedup
Analysis of a dynamically wavelength-routed optical burst switched network architecture
Network flow switching and flow data export
References
College Admissions and the Stability of Marriage
On the self-similar nature of Ethernet traffic (extended version)
Analysis and simulation of a fair queueing algorithm
Input Versus Output Queueing on a Space-Division Packet Switch
Analysis and Simulation of Fair Queueing Algorithm
Related Papers (5)
Frequently Asked Questions (15)
Q2. What are the future works mentioned in the paper "Matching output queueing with a combined input output queued switch1" ?
The authors believe this to be an important area for future research.
Q3. How can GBVOQ be used in conjunction with the DTC strategy?
Like CCF, GBVOQ can be used in conjunction with the DTC strategy to reduce the number of iterations needed to compute a stable matching.
Q4. What is the common conclusion of these studies?
A common conclusion of these studies is that with or 5 one can achieve about 99% throughput when arrivals are independent and identically distributed at each input, and the distribution of packet destinations is uniform across the outputs.
Q5. What is the reason for the CCF insertion policy?
The intuition behind this insertion policy is that a a cell with a small output cushion needs to leave soon (i.e. it is “more critical”), and therefore needs to be delivered to its output sooner than a cell with a larger output cushion.
Q6. What is the CIOQ switch's sequence of departure?
In the CIOQ switch, the sequence in which cells are transferred from their input queues to the output queue is determined by a scheduling algorithm.
Q7. What is the insertion policy for a cell?
“Critical Cells First” (CCF) inserts an arriving cell as far from the head of its input queue as possible, such that the input thread of the cell is not larger than its output cushion.
Q8. What is the common way to queue a packet?
When a packet arrives at an output-queued (OQ) switch, it is immediately placed in a queue that is dedicated to its outgoing line, where it waits until departing from the switch.
Q9. What is the slackness of a cell in a departure phase?
During an arrival phase, the slackness of a cell already in the system can go down by at most F since a new cell with fanout F may get inserted ahead of it.
Q10. What is the way to group incoming cells into virtual output queues?
for emulating a FIFO OQ switch, the authors can group incoming cells into Virtual Output Queues and obtain an upper bound of on the number of cells that need to be considered.
Q11. What is the common practice in high performance LAN switches?
This is common practice in high performance LAN switches and routers; variable length packets are segmented into cells as they arrive, carried across the switch as cells, and reassembled back into packets again before they depart [4][3].
Q12. How does Prabhakar and McKeown show that a CIOQ switch can behave?
They show that a CIOQ switch with a speedup of four can behave identically to a FIFO OQ switch for arbitrary input traffic patterns and switch sizes.
Q13. What is the VOQ that needs to be marked active?
and unfortunately, to determine which VOQs need to be marked active, the authors again need access to global state, namely the output cushion of each cell at the head of a VOQ.
Q14. how can a cell's slackness be determined?
Counting the changes in each of the four phases (arrival, departure, and two scheduling phases), the authors conclude that the slackness of cell c can not decrease from time slot to time slot.
Q15. What is the main difference between output queueing and IQ switching?
This approach is known to maximize the throughput of the switch: so long as no input or output is oversubscribed, the switch is able to support the traffic and the occupancies of queues remain bounded.