1 / 7

GPS Example 2: Arrivals

0. 10. 20. Time. GPS Example 2: Arrivals. GPS Example 2: Arrivals. Eleven Sources. First source gets 0.5. Other 10 sources get 0.05 each. First source sends 11 cells. 2-11 send one each at t=0. S1. S2. p 11 1. S3. S4. S5. S6. S7. S8. S9. S10. S11. GPS Example 2: Service.

verda
Download Presentation

GPS Example 2: Arrivals

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. 0 10 20 Time GPS Example 2: Arrivals GPS Example 2: Arrivals • Eleven Sources. First source gets 0.5. Other 10 sources get 0.05 each. First source sends 11 cells. 2-11 send one each at t=0. S1 S2 p111 S3 S4 S5 S6 S7 S8 S9 S10 S11

  2. GPS Example 2: Service • Each cell of the first source takes 2 units of time. Sources 2-11 take 20 units each. S1 p111 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 0 10 20 Time

  3. WFQ: Service • Packets finish at the same time or earlier than GPS. Some packets finish much earlier. Long period of no service Unfair. S1 S2 p111 S3 S4 S5 S6 S7 S8 S9 S10 S11 0 10 20 Time

  4. Worst Case Fair Weighted Fair Queuing (WF2Q ) • WF2Q fixes the unfairness problem in WFQ. • WFQ: Among packets waiting in the system, pick one that will finish service first under GPS. • WF2Q: Among packets waiting in the system that have started service under GPS, select one that will finish first under GPS. • WF2Q provides service close to GPS (difference in packet service time bounded by max. packet size). • WF2Q+ is an simpler implementation of WF2Q • Refs: Jon Bennett, Hui Zhang.

  5. WF2Q: Service S1 p111 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 0 10 20 Time

  6. Stochastic Fair Queueing (SFQ) • Allows # of queues less than the number of flows • Hash to queues  O(1) time for enqueuing • Packets are serviced in round-robin • If buffers full, packet in the longest queue is pushed out • Known Results: O(1) operation. • # of queues needs to be a small multiple of # of ACTIVE flows rather than total number of flows • Known Problems: Flows hashing to the same queue not isolated • Refs: McKenney, Shreedhar and Varghese

  7. Deficit Round Robin (DRR) • Use hashing to assign flows to queues • Per-Queue Quota, Per-Queue Credit (called deficit) • Send packet if Credit  packet size • After service: Credit = Credit-packet size • Known Results: Fair. O(1). • Latency depends upon frame size = sum of max packet sizes = fn of # of flows • Ref: Shreedhar and Varghese

More Related