THE DEFORESTATION OF L2

Transcription

1 THE DEFORESTATION OF L2 James McCauley, Mingjie Zhao, Ethan J. Jackson, Barath Raghavan, Sylvia Ratnasamy, Scott Shenker UC Berkeley, UESTC, and ICSI

2 The Talk What is AXE? Why look at this? How does it work? Really? This actually works? 2

3 The What An redesign of L2 to replace Ethernet and Spanning Tree Protocol (and its variants) Targets are normal enterprise networks, machine rooms, small private DCs Not the Googles, Microsofts, Rackspaces Not networks with incredibly highly utilization Not managed by a full-time team of experts 3

4 The What: Goals Plug-and-play If not, might as well just use L3 Use all links for shortest paths Number one shortcoming of STP Fast recovery from failure Number two shortcoming of STP? 4

5 The What: Goals Plug-and-play If not, might as well just use L3 Use all links for shortest paths Number one shortcoming of STP Fast Packet-timescale recovery from failure Number two shortcoming of STP? 5

6 The What: Assumptions Failure detection can be fast Not traditionally the bottleneck Control plane hellos were sufficient Need interrupt-driven LFS, BFD, etc. There s a market for flood-and-learn L2 Flooding/learn has security implications No heavy unidirectional traffic No multi-access links Everything is point-to-point 6

7 The Why: Is L2 still a problem? Still many largely-unmanaged, small/med L2 networks! Two in our building in Berkeley! There have been a few interesting developments SPB, TRILL, SEATTLE, etc. Provide various tradeoffs AXE attempts to strike a different balance Focus on two key problems Keeping things as simple as possible (no control plane) 7

8 The Why: Context Plug-and-play Shortest Paths Fast Recovery No Control Plane STP No STP (Tree) TRILL/SPB IP (L3) Custom AXE? 8

9 The How: Extend Ethernet Basic flood/learn Ethernet When you see a packet: learn When you don t know what to do: flood But AXE does not need a tree to deal with loops Means flooding works for handling failures too (because alternate paths are immediately available) Means that flood/learn finds short paths (because you haven t removed links) 9

10 The How: Treeless flooding How do you get around the loop problem? Duplicate-packet-detection Multiple ways of doing it Our focus: hash-based deduplication filter In short: hash table where you replace upon collision Straightforward Amenable to hardware/p4 implementation 10

11 The How: What changes? Learning is more subtle Source address seen on multiple ports Packets may even be going backwards! Responding to failures is more subtle Means we have to unlearn (outdated) state 11

12 The How: Extend Header Extend the packet header between switches Nonce (per-switch sequence number) Used for packet deduplication Hop count Influences learning, also protects from loops Flooded flag: F Tracks whether a packet is being flooded Learnable flag: L Tracks whether packet can be learned from 12

13 The How: Separate queues Switches have flood queue and normal queue The Flooded flag in the header determines which Flood queue has higher priority and is shorter Normal queue sized normally Intuition: Delivering floods quickly stops flooding quickly Deduplication only applies to floods, keeping fewer floods in flight makes dedup easier 13

14 The How: Overview Extend packet header Nonce, Hop Count, Flooded / Learnable flags Learning/Unlearning Phase May learn port and HC to src May unlearn path to dst if trouble was observed Output Phase If packet is a duplicate: drop If unknown-dst/path-failed/already-flooding: flood Otherwise forward according to table 14

15 The How: Example A sends a packet to B (L:True) Destination B unknown; packet flooded from first hop (F:True) All switches learn how to reach A B sends to A (L:True) Direct path following table entries to A (F:False) Switches along path learn how to reach B Link fails A sends another packet to B (L:True) Follows along path (F:False).. until it hits failure (L:False F:True) Switch floods packet out all ports (even backwards) Flooded packet reaches B (Successful delivery!) Another duplicate of flooded packet reaches A s first hop; unlearn B A sends another packet to B (L:True) Destination B unknown; packet flooded from first hop (F:True) All switches learn how to reach A again 15

16 Really? Preliminaries How much flooding do failures cause? Campus Cluster 0.00% 0.02% 0.04% 0.06% 0.08% 0.10% 0.12% 0.14% 0.16% 0.18% How big does the deduplication filter need to be? Less than 1,000 entries in our simulations Does it recover from overload? Yes* 16

17 Really? Overview Thinking back to that matrix We want plug and play We to support shortest paths using all links We don t want to have a control plane Packet-timescale recovery from failures 17

18 Really? Failure benchmark Omniscient, randomized, shortest-path routing Failure Adjustable delay Fix routes Delay of zero is optimal routing / an upper bound Nonzero delay meant to roughly simulate OSPF, IS-IS, TRILL, SPB, etc... without needing to model each one in detail Random shortest-cost tree rooted at each destination Note: we don t compare ourselves to STP at all 18

19 Really? Failure recovery - UDP Send traffic on network with high failure rate Metric is unnecessary delivery failures packets that weren t delivered even though optimal routing could have delivered them AXE has no unnecessary delivery failures 160ms 0.27% 80ms 0.19% 40ms 0.13% 20ms 0.08% 10ms 0.05% 5ms 0.03% AXE 0.00% 0.00% 0.05% 0.10% 0.15% 0.20% 0.25% 0.30% Delivery Failures (Campus) 160ms 0.021% 80ms 0.011% 40ms 0.008% 20ms 0.005% 10ms 0.003% 5ms 0.002% AXE 0.000% 0.000% 0.005% 0.010% 0.015% 0.020% 0.025% Delivery Failures (Cluster) 19

20 Really? Failure recovery - TCP 160ms 0.25% 80ms 0.11% 40ms 0.14% 20ms 0.06% 10ms 0.01% 5ms 0% AXE 0% 0.0% 0.1% 0.1% 0.2% 0.2% 0.3% 0.3% Delayed FCTs (Campus) Similar setup, but with TCP Metric is number of flows with significantly worse FCT than optimal routing 160ms 0.036% 80ms 40ms 20ms 10ms 0.010% 0.007% 0.015% 0.020% AXE has no significantly worse FCTs 5ms 0.002% AXE 0% 0.00% 0.01% 0.01% 0.02% 0.02% 0.03% 0.03% 0.04% 0.04% Delayed FCTs (Cluster) 20

21 The End: Not Mentioned Here Multicast AXE On any change (failure; join), flood+dedup and prune Flooded packets have all data needed to build tree AXE with Hedera Use AXE for mice & recovery Centralized SDN routing for elephant flows P4 implementation AXE is expressible in P4 Performance on real hardware is open question 21

22 THE END 22