Empirical Study - Multimedia Networking - Lecture Slides, Slides of Computer Science

Download Empirical Study - Multimedia Networking - Lecture Slides and more Slides Computer Science in PDF only on Docsity! An Empirical Study of Delay Jitter Management Policies Docsity.com Introduction • Want to support interactive audio • “Last mile” is LAN (including bridges, hubs) to desktop – Study that – (Me: 1995 LANs looked a lot like today’s WANs) • Transition times vary, causing gaps in playout – Can ameliorate with display queue (buffer) Docsity.com This Paper • Evaluates 3 policies for managing display queue – I-policy, E-policy from [NK92] • (I is for late data ignored, E is for expand time) – Queue Monitoring from this paper • Empirical study – Audioconference on a LAN – Capture traces • Simulator to compute delay and gaps Docsity.com Outline • Introduction (done) • The I- and E-policies (next) • The Queue Monitoring policy • Evaluation • The Study • Summary Docsity.com The Effect of Delay Jitter • If display latency worse than largest end-to- end latency, then no gaps – (When is this not what we want?) • Playout with low latency and some gaps preferable to high-latency and no gaps • What if a frame arrives after its playout time? • Two choices: – I-Policy – single fixed latency, so discard – E-Policy – late frames always displayed, so expand playout time Docsity.com I-Policy (2) One event, but latency still low (e, f, g, …) Docsity.com a E- Policy (2) —— Y SSS * Display Queue Len oth 3 Queue 9 rength One event, r LLL | latency 23 4 567 8 9 10 higher Display Initiation Time Display Latency 3 ec dee if ¢g Time 92 (in frames) a p 1 0 — | | — —| — 2 3 4 5 6 7 8 Qg9 10 : rene " Docsity.com Display Initiation Time Policy Summary • Display latency chosen implicitly with E-policy • Choose it explicitly with I-policy • What is the right display latency amount? – Depends on application • Example: surgeon viewing operation vs. televised lecture – Depends on network and machines • Can vary across a long run • So, need a policy that allows display latency to be chosen dynamically Docsity.com Monitor the Queue • Measuring the end-to-end latency is difficult because needs synchronized clocks • Instead, observe length of display queue over time – If end-to-end delay constant, queue size will remain the same – If end-to-end delay increases, queue shrinks – If end-to-end delay decreases, queue expands • If queue length > 2 for some time, can reduce queue (hopefully) without causing a gap – “some time” is parameter, n, in frame times – Implement with counters for each of m frames in queue – If any m times > n, discard frame and reset • (keep queue at least 2) – Use QM-120 as default (about 2 seconds) Docsity.com Outline • Introduction (done) • The I- and E-policies (done) • The Queue Monitoring policy (done) • Evaluation (next) • The Study • Summary Docsity.com Comparing Policies • If A has lower latency and gaps than B, then A is better • If A lower latency, but A higher gaps than which is better? – Depends upon • relative amounts • resolution • application requirements – Few standards Docsity.com The Study • Run videoconference – Use audio only • Record end-to-end delay • Input into simulator to evaluate policy Docsity.com Videoconference • Built at UNC • Runs on IBM PS/2 • Uses UDP • IBM-Intel ActionMedia 750 – 30 fps, 256x240, 8-bit color (6-8 k frames) – Audio 60 fps, 128 kb/second into 16.5ms frames (266 byte packets) Docsity.com Network • 10 Mb Ethernets and 16 Mb token rings • 400 Unix workstations and Macs • NFS and AFS • Send machine  token-ring  gateway  department Ethernet  bridge  department Ethernet  gateway  token- ring  display machine Docsity.com 100000 7 10000 ® 1000 = fg L400 5 a 2 E S Zz 1 o4 100000 4 10000 4 wo 4 oO 1000 = © 100 6 . oO oO — 104 5} Zz 1 o4 Two Example Runs Low jitter 100 150 200 250 High jitter 100 150 200 250 End-To-End Delays (ms.) Docsity.com Results QM-120 better than I-2 for all but 11 (I-2 has gap per 2 seconds vs per 11 seconds) Docsity.com Results Better than I-3 for all but 15 Latency of QM-120 better than that of I-3 Better than E for low jitter runs Docsity.com Results g Run QM (30) QM (60) QM (120) QM (600) QM (3600) [120] 120]120] 120 Latency Gaps] Latency Gaps] Latency Gaps] Latency Gaps] Latency Gaps] vs. | vs. | vs.| vs. ms. /min,| ms. /min.| ms. /min.| ms. /min.| ms. ‘min. | 30 | 60 [600] 3600 1 64 0.3 65 0.3 66 0.3 3 0.3 75 0.2 7,0] 070 0 2 65 0.7 65 0.7 66 0.6 66 0.6 67 0.6 70] 070 0 3 67 1.7 67 4 68 1.4 74 1.4 103 Ll Jo] o]7o + 4 65 0.6 65 0.6 65 0.6 69 0.6 83 0.670] 07 0 + 5 67 0.5 68 0.5 68 0.5 69 0.5 81 05 70] 070 0 6 70 0.5 70 0.5 70 0.5 70 0.5 76 04,0] 070 0 7 70 2.3 71 1.9 72 1.9 77 1.7 95 L4],o0] 070 + 8 68 2.0 70 1.5 75 1.3 83 1.0 97 Lo ],o]oaqTo + 9 77 13.1 83 9.0 87 7.6 102 4.9 117 3.0 7+ ]+ ] - 10 72 6.6 75 5.0 78 3.9 89 1.6 98 Lo }+]+]- 11 72 8.3 76 6.3 83 4.8 98 3.4 124 L7 |] +] + 12 72 5.3 74 3.3 76 2.7 86 1.9 103 12 ]+ ]o]7o0 13 69 3.5 70 2.7 72 2.1 82 1.4 91 Lo }+]oypo 14 74 6.7 76 6.0 80 3.9 92 1.8 99 12 ],+ ]+]- Comments? Docsity.com Summary • QM-600 is the best relative to QM-120 • QM-120 better than all the others • (Me, what about in between? Should be optimal for each setting.) • Also, – QM-3600 similar to E-policy – QM-30 and QM-60 similar to I-2 Docsity.com Decay Thresholds • Want to converge slowly to lowest latency • Base threshold for queue length of 2 • Decay factor for other queue lengths • Base of 3600, decay of 2 would have: – 3600 for 3, 1800 for 4, 900 for 5 … Docsity.com Summary • Will always be delay – From network or OS or … • Need to adjust queue latency – QM-(600,2) is the best, QM-120 almost as good • Queue monitoring can be effective – 35-40 ms delay, variation up to 200ms, even 80 ms when quiet • Run 3 Best vs. E – E: 140ms, .9 gaps/min – QM-(600,2): 68ms, 1.4 gaps/min • Run 24 Best vs. I – I: 93 ms, 15 gaps/min – QM-(600,2): 90 ms, 4 gaps/min • QM is flexible, can be tuned to app or user Docsity.com Future Work? Docsity.com Future Work • Compare against I-policy where threshold changes each talkspurt • Compare using different metrics, say that combine latency and gaps or looks at distribution – PQ studies to measure tradeoffs • Larger networks • Combine with repair • Other decay strategies for QM Docsity.com