<< Chapter < Page
  Ueh network   Page 1 / 1
Chapter >> Page >
(Blank Abstract)

A polling network is a computer communications network that uses polling to control access to the network. Each node or station on the network is given exclusive access to the networkin a predetermined order. Permission to transmit on the network is passed from station to station using a special message called a poll . Polling may be centralized (often called hub polling ) or decentralized ( distributed ). In hub polling, the polling order is maintained by a single central station or hub . When a station finishes its turn transmitting, it sends a message tothe hub, which then forwards the poll to the next station in the polling sequence. In a decentralized polling scheme, each stationknows its successor in the polling sequence and send the poll directly to that station. To simplify matters, we will assume adistributed polling scheme.

The analysis of a polling network uses the results of the analysis of an M/G/1 queue with vacations. Each vacationcorresponds to the transfer of the poll from one station to the next in the polling cycle. We divide time into alternatingtypes of intervals: polling intervals , during which the poll is transferred between stations, and transmission intervals , during which the station with the poll transmits packets.

Polling networks come in three flavors: gated, exhaustive, and partially gated. In a gated system, each station is allowed to transmit only those packets that arrived prior tothe start of the poll interval (i.e., prior to the start of the vacation preceding the station's use of the network). An exhaustive scheme allows a station to transmit any packets that arrive before it transfers the poll to the nextstation. A partially gated network allows stations to transfer all packets that arrive by the time the poll does.Polling networks will typically be partially gated or exhaustive, not gated.

We assume that arrivals at each of the m stations are independent Poisson processes with rate λ m . Note that "arrival" refers to a message arriving from the "outside world" to a station in order to be transmittedover the network; it does not mean the arrival of a message that has been transmitted over the network.

Gated system, m=1

We define the following notation:

  • X - mean packet length in seconds
  • X 2 - second principal moment of the packet length distribution
  • V r - r.v. for length of r th polling interval. V r are independent and identically distributed.
  • V V r
  • V 2 V r 2
  • R i - r.v. for residual time data packet i must wait in queue until end of current packettransmission or polling interval
  • V r ( i ) - r.v. for length of polling interval for data packet i
  • ρ λ X
A packet which arrives in a gated system with one station must first wait the residual service time for the packet currentlybeing transmitted or the residual length of the poll transfer time, depending on when it arrives. It must then wait untilall of the packets queued for transmission at its arrival have been serviced. Finally, it must wait until the next polltransfer is finished (note that in this simplistic system, the station sends the poll to itself). W i R i N i X V r ( i ) The time-average for the residual service time can be obtained as it was for the M/G/1 queue with vacations. R i λ X 2 2 1 ρ V 2 2 V Also as before, i N i X i λ W i X ρ W i V r ( i ) V W λ X 2 2 1 ρ V 2 2 V ρ W V λ X 2 2 1 ρ V 2 2 V V 1 ρ If V i A , a constant for all i , this can be simplified to W λ X 2 2 1 ρ A 2 2 A A 1 ρ λ X 2 2 1 ρ A 1 ρ 2 A 2 1 ρ λ X 2 2 1 ρ A 2 3 ρ 1 ρ

M>1

The case of one station has relatively little practical application in computer networks, but the analysis does serveas a convenient starting point for m 1 stations. We first need to define some additional notation:

  • Y i - r.v. for combined length of all of the whole polling intervals during which packet i must wait
  • N i - r.v. for total number of packets that must be transmitted after the arrival of packet i and before i is transmitted (not including any packet in service when i arrives)
  • R i - r.v. for residual time for the packet or poll in progress
  • Y i Y i
A packet must wait
  • while the packet transmission or poll interval underway at its arrival finishes;
  • for all packets which arrived before it but which had not been serviced yet to be transmitted;
  • for the time required to transfer the polls from station to station until the transmission interval in which thepacket will be transferred starts.
W i R i N i X Y i As before R R i λ X 2 2 1 ρ r 0 m 1 V r 2 2 r 0 m 1 V r Each packet transmitted before i has an average transmission time X . N i is not the number of packets ahead of i in the queue at its arrival, since packets might arrive at other stations after i arrives, but actually be transmitted before i because of the polling cycle. However, by Little's Theorem, i N i X λ W X ρ W Consequently, W R ρ W Y R Y 1 ρ Y depends on the flavor (gated, partially gated, or exhaustive) of polling network. We willlook at each.

Exhaustive system

α j k is the expected value of Y i given that packet i arrives in user j 's polling or data interval and belongs to user j k m . α j k 0 k 0 V ( j + 1 ) mod m V ( j + k ) mod m k 0 We first remove the condition that the packet belongs to station j k m by assuming that a packet belongs to a particular station with probability 1 m for all stations. The expected value of Y i given that packet i arrives in user j 's polling or data interval is given by 1 m k 1 m 1 α j k k 1 m 1 m k m V ( j + k ) mod m Since all users are identical, they have equal average length data intervals in steady-state, and the steady-stateprobability that a packet arrives in a particular user's data interval is ρ m . Similarly, the probability that a packet arrives during a particular user's polling interval is 1 ρ V r k 0 m 1 V k .

Y r 0 m 1 ρ m 1 ρ V r k 0 m 1 V k j 1 m 1 m r m V ( r + j ) mod m ρ m r 0 m 1 j 1 m 1 m j m V ( r + j ) mod m 1 ρ k 0 m 1 V k r 0 m 1 V r j 1 m 1 m j m V ( r + j ) mod m ρ m r 0 m 1 m j m j 1 m 1 V ( r + j ) mod m 1 ρ k 0 m 1 V k r 0 m 1 j 1 m 1 m j m V r V ( r + j ) mod m
r 0 m 1 j 1 m 1 m j m V r V ( r + j ) mod m r 0 m 1 j 0 m 1 m j m V r V ( r + j ) mod m r 0 m 1 V r 2 1 2 r 0 m 1 V r 2 r 0 m 1 V r 2
The mean polling interval length averaged over all users is V 1 m r 0 m 1 V r Thus
Y ρ m j 1 m 1 m j V 1 ρ 2 m V m V 2 r 0 m 1 V r 2 ρ V m j 1 m 1 m j 1 ρ m V 2 1 ρ r 0 m 1 V r 2 2 m V ρ V m 1 m 1 ρ m V 2 1 ρ r 0 m 1 V r 2 2 m V m ρ V 2 1 ρ r 0 m 1 V r 2 2 m V
Substituting these expressions for R and V into the equation for W ,
W λ X 2 2 1 ρ r 0 m 1 V r 2 2 m V m ρ V 2 1 ρ r 0 m 1 V r 2 2 m V λ X 2 2 1 ρ m ρ V 2 1 ρ r 0 m 1 V r 2 V r 2 2 m V
Thus, for exhaustive gating, W λ X 2 2 1 ρ m ρ V 2 1 ρ σ V 2 2 V

Partially gated system

In the partially gated system, a packet that arrives during a user's own data interval is delayed by an additional m V on average, and this occurs with probability ρ m , thus increasing Y by ρ V compared to the exhaustive case. W λ X 2 2 1 ρ m ρ V 2 1 ρ σ V 2 2 V

Fully gated system

If the system is fully gated, then a packet that arrives during a user's own polling interval is also delayed by anaverage of m V , and this is in addition to the extra delay incurred in the partially gated case. The probability of thisoccurring is 1 ρ m , thus increasing Y by 1 ρ V compared to the partially gated case. W λ X 2 2 1 ρ m 2 ρ V 2 1 ρ σ V 2 2 V

Get Jobilize Job Search Mobile App in your pocket Now!

Get it on Google Play Download on the App Store Now




Source:  OpenStax, Ueh network. OpenStax CNX. Jul 20, 2006 Download for free at http://cnx.org/content/col10367/1.1
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'Ueh network' conversation and receive update notifications?

Ask