1 / 17

SRED: Stabilized RED

SRED: Stabilized RED. T. Ott, T.V. Lakshman, L. Wong Presented by King-Shan Lui. Diff. With RED. SRED estimates number of active flows Misbehaving flows can be identified without keeping per-flow state Drop probabilities are adjusted according to number of active flows

denton
Download Presentation

SRED: Stabilized RED

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. SRED: Stabilized RED T. Ott, T.V. Lakshman, L. Wong Presented by King-Shan Lui

  2. Diff. With RED • SRED estimates number of active flows • Misbehaving flows can be identified without keeping per-flow state • Drop probabilities are adjusted according to number of active flows • No computation of average queue length • Assume TCP flows

  3. Main Idea • Number of active flows  number of different flows in the buffer • A misbehaving flow has a lot of packets in the buffer • When a packet arrives, compare it with a packet arrived before. If they belong to the same flow, a hit occurs.

  4. Zombie List • A list of M recently seen flows, zombies • Longer memory than the buffer alone • Information for each zombie: • Count: number of packets of this zombie received • timestamp: arrival time of the most recently received packet

  5. Zombie List Operations • Zombie list is not full • Insert the flow with count = 0, timestamp = ta • Zombie list is full • Randomly pick a zombie • Hit: count += 1, timestamp = ta • No hit: with prob. p that the zombie is replaced • The arrived packet may be dropped no matter there was a hit or not

  6. Hits & Number of Active Flows • Zombie list loses memory once every M/p packets • Few active flows  more hits • Misbehaving flows cause more hits than well-behaved flows

  7. Hit Frequency • P(t) – hit frequency around the time of the tth packet arrives at the buffer • Hit(t) = 1 when hit; Hit(t) = 0, otherwise • P(t) = (1 – a)P(t – 1) + a*Hit(t) • Proposition: P(t)-1 is a good estimate for the effective number of active flows

  8. Proposition Argument • P(arrival packet belongs to flow i) = pi • P(Hit(t)=1) = Spi2 • 1/N  Spi2  1 • Symmetric case: N flows, pi = 1/N • P(t) = 1/N (exact estimate) • Asymmetric case: infinite flows, pi = 2-i • P(t) = 3/16 (effective number of active flows)

  9. Simple Stabilized RED • Target buffer occupation – Q0 • Set a drop probability – p • Square root law: congestion window of each flow, cwnd  p-1/2 • Sum of N congestion windows – N * p-1/2 • Q0 = N*p-1/2 p = (N/Q0)2 • p is proportional to N2

  10. Simple Stablized RED (cont.) • Buffer capacity – B • Current buffer size – q • pzap= psred(q) * • psred(q) = pmaxif 1/3*B  q < B = ¼ * pmax if 1/6*B  q < 1/3*B = 0 if 0  q < 1/6*B

  11. psred 1 pmax q B psred(q) • Depends on current q, not history of q • Three levels • Ratio 4: halving the • congestion windows

  12. pzap • When number of flows  256 pzap ~ psred/65356 * (number of flows)2 • When number of flows > 256 pzap = psred • Avoid pzapbecomes too large • pzap depends on q and P(t)

  13. Full SRED Increase the drop probabilities of misbehaving flows

  14. Simulation Results

  15. Simulation Results • SRED stablilizes the buffer occupancy when N  256 as q is independent of N • q increases slightly when N = 1000 • Buffer occupation almost never decreases below B/6 • could narrow the band where psred(q) = pmax/4

  16. Contributions • SRED provides a mechanism to estimate number of active flows and identify misbehaving flows • SRED controls buffer occupancy by adjusting drop probabilities using estimated number of active flows

  17. Remaining Issues • Many parameters to be tuned: psred, pmax, a, M, p, magic 256. • Extra storage: Zombie list vs. per-flow state • M ~ 1000 > 256

More Related