Cheatsheet: Difference between revisions

Content added Content deleted
Line 208: Line 208:


[[File:Mtu mss.png|center]]
[[File:Mtu mss.png|center]]

⚫
* '''RTO:''' Four ACKs acknowledging the same packet, which are not piggybacked on data and do not change the receiver's advertised window.

⚫
*Fast Retransmission
⚫
- If RTO has a larger value
⚫
- If sender receives four acknowledgments with same value (three duplicates)
⚫
- Segment expected by all of these Ack is resent immediately

*Fast Recovery:
⚫
-
-


*Congestion Control
*Congestion Control

;Slow Start - Exponential Increase
;Slow Start - Exponential Increase
- Sender starts with cwnd = 1 MSS, Size increases 1 MSS each time one Ack arrives, Increases the rate exponentially(1,2,4,8....) until a threshold is reached
- Sender starts with cwnd = 1 MSS, Size increases 1 MSS each time one Ack arrives, Increases the rate exponentially(1,2,4,8....) until a threshold is reached
Line 236: Line 248:
* Behavior of Tahoe and Reno differ primarily in how they react to duplicate ACKs.
* Behavior of Tahoe and Reno differ primarily in how they react to duplicate ACKs.


<center>
Reno:
If three duplicate ACKs are received, Reno will
Skip the slow start phase by instead halving the congestion window (instead of setting it to 1 MSS like Tahoe),
Setting the slow start threshold equal to the new congestion window
Enter a phase called fast recovery.

* In both Tahoe and Reno, if an ACK times out (RTO timeout), slow start is used, and both algorithms reduce congestion window to 1 MSS.
⚫
* Four ACKs acknowledging the same packet, which are not piggybacked on data and do not change the receiver's advertised window.


{| class="wikitable"
{| class="wikitable"
|-
|-
! Event !! Tahoe !! Reno
! Event !! Tahoe !! Reno
|-
|-
| 3 Dup Acks || performs a fast retransmit<br> Sets the slow start threshold to half of the current congestion window<br>Reduces the congestion window to 1 MSS<br>Resets to slow start state|| perform a fast retransmit
| 3 Dup Acks || Performs a fast retransmit<br>Sets the slow start threshold to half of the current congestion window<br>Reduces the congestion window to 1 MSS<br>Resets to slow start state || Perform a fast retransmit<br>Skip the slow start phase by instead halving the congestion window<br>(instead of setting it to 1 MSS like Tahoe)<br>Setting the slow start threshold equal to the new congestion window<br>Enter a phase called fast recovery.
|-
|-
| RTO (Ack time out) || Slow start is used<br>Reduce congestion window to 1 MSS || Slow start is used<br>Reduce congestion window to 1 MSS
| RTOs || Example || Example
⚫
|-
| Example || Example || Example
|}
|}
</center>

----


*Silly Window Syndrome: Sender creates data slowly or Receiver consumes slowly or both.
*Silly Window Syndrome: Sender creates data slowly or Receiver consumes slowly or both.
Line 268: Line 268:
- '''Delayed Acknowledgment''': Segment not acknowledged immediately, Sender TCP does not slide its window, reduces traffic, sender may unnecessarily retransmit, Not delay more than 500 ms.
- '''Delayed Acknowledgment''': Segment not acknowledged immediately, Sender TCP does not slide its window, reduces traffic, sender may unnecessarily retransmit, Not delay more than 500 ms.


⚫
*Fast Retransmission
⚫
- If RTO has a larger value
⚫
- If sender receives four acknowledgments with same value (three duplicates)
⚫
- Segment expected by all of these Ack is resent immediately


*Persistence Timer
*Persistence Timer
Retrieved from "https://aman.awiki.org/wiki/Cheatsheet"