Download presentation

Presentation is loading. Please wait.

Published byKathleen Pembleton Modified over 2 years ago

1
Quantifying Overprovisioning vs. Class-of-Service: Informing the Net Neutrality Debate Murat Yuksel (University of Nevada – Reno) yuksem@cse.unr.edu K. K. Ramakrishnan (AT&T Labs Research) kkrama@research.att.com Shiv Kalyanaraman (IBM Research India) shivkumar-k@in.ibm.com Joseph D. Houle (AT&T) jdhoule@att.com Rita Sadhvani (Verizon Wireless) rita.sadhvani@verizonwireless.com 1

2
Motivation: Thick (Over-provisioned) or Thin (Engineered) Pipes? Thin: How to deal with bursts/overload? And meet premium SLAs… ! Thick: Cost of overprovisioning? Can this commodity model break even? 04000080000 10000 0 rate time [Jim Roberts et al.] Media-rich applications require performance guarantees: –e.g.: VoIP requires <300ms round-trip delay, <1% loss How to respond to these application needs? –CoS approach: provide priority (i.e. higher class) to premium traffic –Classless (best-effort) service approach: over-provision the capacity Question: How much extra capacity does the classless service require to match the performance of the higher class (premium) service in the CoS approach? 04000080000 10000 0 rate time 2

3
Two Approaches: CoS vs. Classless Premium BE DD CoS Link (differentiated) D Prem = g D BE =(1-g) D D GIVEN: D, D and a performance target (i.e. t target or p target ) FIND: The minimum N that gives the same performance as in the premium class of the CoS case? N =? Classless Link (neutral) BE Scheduling (e.g. priority) 3

4
REC: Required Extra Capacity REC= - = N - D (rate) = 100( N / D – 1)(%) How to quantify REC? 4

5
Analytical Link Model: Poisson traffic Assume: –Poisson traffic, Exponential packet lengths for traffic in each class i.e. Premium class traffic is Poisson with g D Best-effort class traffic is Poisson with (1-g) D –The aggregate traffic for the neutral link is also Poisson with rate D 5 Both the performance target and the REC can be expressed in terms of two key parameters: (i) ρ – utilization, (ii) g – proportion of premium traffic.

6
More Bursty Traffic: MMPP MMPP = Markov-Modulated Poisson Process –Easy to do the math… –Simplest MMPP is of two states. MMPP traffic with mean D –Traffic w/ equivalent rate to the neutral case, but w/ more burstiness. 1 2 Higher r means more bursty traffic. 6

7
Simulated Link Model: Delay MMPP/M/1 model a=0.5, r=4 If packet size is 1KB and the CoS link is D = 1Gb/s: 5,000packets of delay = 40.1ms Surface color shows the performance target. REC can be quite high even for very small g and medium utilizations. 7

8
Simulated Link Model: Delay MMPP/M/1 model a=0.5, r=4 REC increases as link utilization increases REC is large even for small proportion of premium traffic 8 Can be drawn in multiple 2-d graphs

9
Simulated Link Model: Loss MMPP/M/1/K model The graphs are generic for various buffer sizes. An example: For a 10Mb/s link carrying 1KB packets: K = ~15pkts 25ms buffer time K = ~60pkts 100ms buffer time 9 REC for the same performance target decreases as buffer size increases

10
Simulated Link Model: LRD Traffic 10 Internet traffic : known to be LRD with Hurst parameter value between 0.7 and 0.9. REC for Hurst=0.75 is significantly higher than our 2-state MMPP model results. We also observed that REC increases as Hurst value increases towards 0.9. DELAY – LRD/D/1 LOSS – LRD/D/1/K Also looked at closed-loop traffic - many TCP flows - and observed similar trends. We further looked at the case when Premium traffic is CBR and BE is TCP, and this increased REC further.

11
Network Model Steps to calculate network REC (NREC): –Step 1: Construct the routing matrix R FxL based on shortest path Run Dijkstra’s algorithm on the topology matrices A NxN and W NxN –Step 2: Form the traffic vector Fx1 from T NxN –Step 3: Calculate the traffic load on each link: R T = Q –Step 4: Check the feasibility of the traffic load and routing For any link –If link capacity is less than the traffic load (e.g. C < Q) then update T accordingly and go to Step 2. –Step 5: Calculate the required per-link REC (i.e. N - D ) by using Q I as the traffic rate D for Ith link, and the performance goal p target or t target. Used Rocketfuel topologies for A NxN and W NxN. Used gravity model for T NxN. Made a look-up to the simulated link model results. 11

12
NREC: Two ways to calculate Steps to calculate network REC (NREC) (cont’d): –Step 6: Calculate the NREC by averaging the per-link RECs from Step 5. We calculated NRECs for the Rocketfuel topologies: –Used the MMPP link model (a=0.5 and r=4) or the LRD link model (H=0.75) – Much more conservative than real or TCP traffic –Assumed K=100ms buffer time –Only report Sprintlink, as the other topologies gave higher REC values 12 total extra capacity needed on the whole network average extra capacity needed on each link

13
NREC for Sprintlink: G2G Delay Solid lines are NREC I and dashed lines are NREC A 13 NREC can be much higher than 100% for a network operating with 60% utilization. 10ms queueing delay target for VoIP may require large REC values.

14
NREC for Sprintlink: G2G Loss Solid lines are NREC I and dashed lines are NREC A 14 NREC can be much higher than 1000% even for a network operating with 40% utilization. 0.1% loss target may require large REC values.

15
Summary A framework to study REC for delay or loss being the performance target. Link model –REC grows when: traffic becomes more bursty the utilization of the CoS link becomes higher the performance target becomes tighter the fraction g of the Premium class traffic becomes smaller –Closed-loop (e.g., TCP) or LRD traffic further increases REC Network model: –For legacy g2g performance targets, REC ranges from 50% to over 100% as g reduces below 0.5 and the utilization goes up to 60%. Future trends/work: –The performance targets will keep becoming tighter. REC is high perpetually – not just today, but in future also.. –The value of g is a crucial factor. Small g does not necessary favor a classless network. –Further research should estimate the actual costs of CoS and classless designs, as scheduling & management complexity need to be considered. 15

16
Thank you! THE END 16

Similar presentations

OK

1 Chapters 9 Self-SimilarTraffic. Chapter 9 – Self-Similar Traffic 2 Introduction- Motivation Validity of the queuing models we have studied depends on.

1 Chapters 9 Self-SimilarTraffic. Chapter 9 – Self-Similar Traffic 2 Introduction- Motivation Validity of the queuing models we have studied depends on.

© 2017 SlidePlayer.com Inc.

All rights reserved.

Ads by Google

What does appt only mean one thing Option call put ppt on loop Ppt on operation research simplex method Ppt on immunization schedule in india Download ppt on hazards of mining Ppt on diode as rectifier tube Mp ppt online form 2012 Ppt on soft skills and personality development Download ppt on women empowerment Service marketing ppt on hotel industry