Slide 1

Slide 1 text

Research Paper Introduction #5 “Experiences Evaluating DCTCP” @cafenero_777 2019/09/10

Slide 2

Slide 2 text

$ which • Experiences Evaluating DCTCP • Lawrence Brakmo, Boris Burkov, Greg Leclercq, Murat Mugan • Facebook • Linux Plumbers Conference 2018

Slide 3

Slide 3 text

Agenda • ֓ཁͱಡ΋͏ͱͨ͠ཧ༝ • Abstract • Introduction • Overview • Intra-Rack Test Results • Inter-Rack Test Results • Conclusions • References

Slide 4

Slide 4 text

֓ཁͱಡ΋͏ͱͨ͠ཧ༝ • ֓ཁ • େن໛DC Closӡ༻ͰͷDCTCPύϑΥʔϚϯεൺֱ݁Ռ • Linux kernel΍NIC’s fi rmware, CPU utilizationͷൺֱ • ಡ΋͏ͱͨ͠ཧ༝ • ࣮ࡍͷେن໛Closߏ੒Ͱͷੑೳൺֱ΍ϋϚΔϙΠϯτ͕஌Γ͍ͨͨΊ

Slide 5

Slide 5 text

Introduction • TCP30೥ͷྺ࢙ • fully, fairly, utilize the available bandwidth • ᫔᫓੍ޚ,Van Jacobson 1988 • ଟ͘͸loss-based algorithm • ύέϩε͢Δ·ͰΩϡʔʹஷΊΔ->(tail)ϨΠςϯγ૿Ճʂ • ύέϩεΛݕ஌͢ΔͨΊʹύέϩεΛ”଴ͭ” • ຊ຤స౗ʁ

Slide 6

Slide 6 text

Introduction (Cont.) • Congestion avoidance algorithm • ϩε͢Δ”લ”ʹ᫔᫓ݕ஌ • Ωϡʔͷ੒௕Λ᫔᫓લஹͱͯ͠ݕ஌ • TCP-Vegas, BBR: RTTΛར༻ • DCTCP: ECNΛར༻ • Reno: • ᫔᫓ݕ஌: cwnd 50%ݮʂ • RTTΑΓ΋୹͍᫔᫓΋ݕ஌ͯ͠͠·͏ • DCTCP: • cwndͷݮΒ͠ํΛ޻෉ • RTTຖʹbyteׂ߹ΛτϥοΩϯά • 100%᫔᫓->cwnd 50%ݮ • 50%᫔᫓->cwnd25%ݮ • ࣮ࡍ͸ҠಈฏۉΛ࠾༻ • 100%᫔᫓͔ͭҎલʹ᫔᫓͕ແ͚Ε͹ • -> cwnd 1/32ݮ

Slide 7

Slide 7 text

Overview • Intra-Rack Test • 3 Sender -> 1 Receiver • 1MB & 10KB RPC • DCTCPಈ࡞֬ೝ • ύέϩεݮ, tail latencyݮ, fairness • Inter-Rack Tests • 3 Worker Racks -> 3 Storage Racks • workers read storage • netestoΛ༻͍ͯղੳ • https://github.com/facebook/fbkutils/tree/master/netesto

Slide 8

Slide 8 text

Intra-Rack Tests Results • 1. ᫔᫓੍ޚʹدΒͳ͍ෆެฏੑͷ໰୊ • Server 1: 25%, Server 2: 25%, Server 3: 50% • εΠονόοϑΝઃܭʹґଘ • Work-around: Bu ff er AͷΈ࢖͏͜ͱͰެฏੑ୲อ • 2. ECNΛ࢖ͬͨ৔߹ͷϑϩʔͷެฏੑͷภΓ໰୊ • 25GbpsΛֻ͚Δɻ΋͏Ұͭͷϑϩʔ͸0.5Gbps͔͠ग़ͳ͍ • 60usͱ1.2msͷόΠϞʔμϧ෼෍ʹͳͬͯ͠·ͬͨ • NICϑΝʔϜ΢ΣΞͷ৽ػೳ͕ݪҼ -> ແޮԽ fl ow# RTT։࢝࣌ؒ (ܦա࣌ؒ) ૹ৴όΠτ਺ Ϩʔτ

Slide 9

Slide 9 text

Intra-Rack Tests Results (Cont.) • 3. DCTCPͷʹΑΔߴtail-latency໰୊ • CubicΑΓDCTCPͷํ͕஗͍ɻɻɻ • 2015೥ͷpatchʹىҼ • όάमਖ਼ޙ • 10KBͷੑೳ͕େ෯޲্ • 1MBͷํ͸૬ରతʹੑೳΛऔΒΕͨͨΊɺ
 Cubicͱൺ΂ͯେ͖͍ • RPCαΠζؒͷෆެฏੑ΋ղܾ

Slide 10

Slide 10 text

Inter-Rack Test Results • Worker͕storageΛಡΉϕϯνϚʔΫ • 3rack, 70% of FSW links • DCTCP͕ͱͯ΋ྑ͍݁Ռ • +1% CPUෛՙఔ౓ • 2rack, 99% of FSW links • DCTCPͷdiscardߴ͍ɺCPUෛՙ • ECNϚʔΫ͕63.7%

Slide 11

Slide 11 text

Inter-Rack Test Results (Cont.) • 1rack, 99.9% of FSW links • discards, ࠶ૹଟ͍ɻɻ • ߴෛՙ࣌ͷDCTCPͷޮՌ͸ݶఆతʁ • ਓ޻తͳϕϯνϚʔΫա͗ͨʁ • prodͰಉ͜͡ͱ͕ى͖Δ͔ෆ໌ • CPUෛՙݮ • ͓ͦΒ͘᫔᫓ͷͨΊcwnd͕ݮ -> ECNϚʔΩϯάݮͷͨΊɻ

Slide 12

Slide 12 text

Conclusions • খ͞Ίͳintra-rackͰ΋DCTCPͷॏཁͳൃݟɺमਖ਼Λݟ͔ͭΒΕͨ • ̏೥ӽ͠ͷόά • inter-rackͰ͸แׅతʹύέϩεݮɺlatencyվળΛ؍ଌͰ͖ͨ • DCTCPͰRPCͷlatencyվળ • ࣮ϫʔΫϩʔυͰCPUෛՙ͕໰୊ʹͳΔ͔ߋʹݕূ͕ඞཁ

Slide 13

Slide 13 text

EoP