rp30
play

#RP30 Automated Deployment and Scaling of Named Data Networks in - PowerPoint PPT Presentation

Aug 02, 2023 •268 likes •737 views

#RP30 Automated Deployment and Scaling of Named Data Networks in Cloud Environments Sean Liao Supervisor: Zhiming Zhao 1 Next 20 mins... Some academic(s) arrive to tell us that (once again) they have Fixed the Internet, and (once again) it

  1. #RP30 Automated Deployment and Scaling of Named Data Networks in Cloud Environments Sean Liao Supervisor: Zhiming Zhao 1

  2. Next 20 mins... Some academic(s) arrive to tell us that (once again) they have Fixed the Internet, and (once again) it runs on top of the current actually-working internet, and (once again) if you sign up you can communicate with as many as twelve other computers. n-gate.com in reference to SCION 2

  3. Introduction 3

  4. ENVRI-FAIR ENVironmental Research Infrastructures (ENVRI) connection to European Open Science Cloud (EOSC) 4 https://envri.eu/about-envri-fair/

  5. Data distribution 5

  6. Content Distribution Network 6

  7. Federated Clouds 7

  8. Named Data Networking (NDN) 8

  9. Named Data Networking NSF Future Internet Architecture Program Information distribution network Potential benefits: ● content caching ● network level security of data 9

  10. 1.2.3.4/a/b/c/d 10

  11. Route directly to data IPv4: 1.2.3.4 IPv6: 1:2:3:4:5:6:7:8 NDN: /arbitrary/strings/infinite/address/space 11

  12. 1-1 Request-Response (Interest-Data) 12

  13. In network caching 13

  14. Run it 14

  15. In the Cloud Existing Tools Router: NFD Overlay over IP Link State Routing: NLSR Simplify deployment Dynamic route updates on static network Scalable Connectivity: FCH Find closest hub/gateway 15

  16. Overlay over TCP / UDP Each point is average over 10K requests 16

  17. Processing Overhead Subtract network roundtrip time: 0.174ms 17

  18. Testing notes Performance severely degrades with 50000+ cached / in-flight requests 18

  19. Scaling Up 19

  20. Problem: Growing a node 20

  21. Replace 21

  22. New Node in Front 22

  23. New Node Behind (Chain) 23

  24. New Node Behind (Load Balancing) 24

  25. Growing the Network 25

  26. Problem: New Node 26

  27. Broadcast / Multicast 27

  28. Bootstrap Gossip 28

  29. Central Discovery Server 29

  30. Network Architecture 30

  31. 3 Layer Cache 31

  32. Mesh Network 32

  33. Automation 33

  34. In-band Management 34

  35. Out of Band Management 35

  36. Proof of Concept 36

  37. Necessary Configuration Discovery Server ● Write down its address / Give it a preconfigured address Load Balancer ● Address of Discovery Server ● Write down its address / Give it a preconfigured address Caching Server ● Address of Load Balancer 37

  38. 1 Cache No Cache 2 Caches Increased Path 38

  39. Proof of Concept Performance Scraping state of router through CLI (on a timer) TCP connections propagate routes and updates Coarse grained partitioning of routes 39

  40. Conclusion 40

  41. Does it work? Reuse existing router in a load balancing configuration Minimal configuration, self connecting network 41

  42. Lessons Learned Now Future Rapidly evolving research testbed Some new network based for new ideas on these ideas Things break or are not optimized Or maybe not (IPv6…) 42

  43. Demo: add load balancer (40s) 43

  44. Demo: remove load balancer (16s) 44

  45. Demo: add Cache (44s) 45

  46. Demo: remove Cache (17s) 46

More recommend