Piotr Srebrny 1. Problem statement Packet caching Thesis claims Contributions Related works ...

Piotr Srebrny

1

Problem statementPacket cachingThesis claimsContributionsRelated worksCritical review of claimsConclusionsFuture work

2

Internet is a content distribution network P2P, file hosting, and streaming account for

more than 80% of the Internet traffic (“Internet study”, Ipoque, 2009)

Single source multiple destination transport mechanism is fundamental

At present, Internet does not provide efficient multi-point transport mechanism

3

Server transmitting the same data to multiple destinations is wasting the Internet resources The same data traverses the same path

multiple times

D

C

S

A

BP D P C P B P A

4

The goal of this work is to

remove the Internet redundancy in a minimal invasive way

5

“Datagram routing for internet multicasting”, L. Aguilar, 1984 – explicit list of destinations in the IP header

“Host groups: A multicast extension for datagram internetworks”, D. Cheriton and S. Deering, 1985 – destination address denotes a group of host

“A case for end system multicast”, Y. hua Chu et al., 2000 – application layer multicast

6

Consider two packets A and B that carry the same content and travel the same few hops

P AA

BP P P

7

Consider two packets A and B that carry the same content and travel the same few hops

P BA

BP P P

B P B

8

I. Packet Caching system can achieve near multicast bandwidth savings

II. Packet Caching system requires server support

III. Packet Caching system is incrementally deployable

IV. Packet Caching system preserves fairness in Internet

9

Principles

Feasibility

Environmental impact

10

I. Contribution

11

Network elements: Link

Medium transporting packets Very deterministic Throughput limited in bits per second

Router Switches data packets between links Very unpredictable Throughput limited in packets per second

12

(I) Cache payloads on links

Caching is done on per link basis Cache Management Unit (CMU) removes

payloads that are stored on the link exit Cache Store Unit (CSU) restores payloads

from a local cache

13

Link cache processing must be simple ~72ns to process a minimum size packet on a

10Gbps link Modern memory r/w cycle ~6-20ns

Link cache size must be minimised At present, a link queue is scaled to 250ms of

the link traffic Difficult to build!

14

(II) A source of redundant data must support caching!

Server can transmit packets carrying the same data within a minimum time interval

Server can mark its redundant traffic

Server can provide additional information on packet content

15

Two components of the CacheCast system Server support Distributed infrastructure of small link caches

D

C

S

A

BD C B P A

CMU CSU CMU CSU

16

Packet train Only the first packet carries the payload The remaining packets truncated to the

header

18

Packet train duration time

It is sufficient to hold payload in the CSU for the packet train duration time

What is the maximum packet train duration time?

19

Back-of-the-envelope calculations

~10ms caches are sufficient

20

II. Contribution

21

Two aspects of the CacheCast system

I. Efficiency How much redundancy CacheCast

removes?II.Computational complexity

Can CacheCast be implemented efficiently with the present technology?

22

CacheCast and ‘Perfect multicast’ ‘Perfect multicast’ – delivers data to

multiple destinations without any overhead

CacheCast overheadsI. Unique packet header per destinationII.Finite link cache size resulting in payload

retransmissionsIII.Partial deployment

23

Example

Lm – total amount of multicast linksLu – total amount of unicast links

u

mm L

L1

5,9 mu LL

%449

4

9

51 m

24

CacheCast unicast header part (h) and multicast payload part (p)

Thus:

E.g.: using packets where sp=1436B and sh=64B, CacheCast achieves 96% of the ‘perfect multicast’ efficiency

uph

mpuhCC Lss

LsLs

)(1

u

mm L

L1

p

hmCC s

sr

r

,1

1

25

System efficiency δm for 10ms large caches

26

S

CMU and CSU deployed partially

1 2 3 4 5 6

27

28

Computational complexity may render CacheCast inefficient

Implementations Server support – a Linux system call and

an auxiliary shell command tool

Link cache elements – implemented with Click Modular Router Software as processing elements

30

New system call msend()

msend() compared with the standard send()system call

31

Server transmitting to100 destinations using Loop of send() sys. calls A single msend() sys. call

msend() system call outperforms the standard send() system call when transmitting to multiple destinations

32

Click Modular Router Software CMU and CSU implemented as

processing elementsCacheCast router

33

Due to CSU and CMU elements CacheCast router cannot forward packet trains at line rate

34

When compared with a standard router CacheCast router can forward more data

35

Testbed setup

Clients from machines A and B gradually connect to server S

36

Original paraslash server can only handle 74 clients

CacheCast paraslash server can handle 1020 clients and more depending on the chunk size

Server load is reduced when using large chunks

37

III. Contribution

38

Internet congestion avoidance relies on communicating end-points that adjust transmission rate to the network conditions

CacheCast transparently removes redundancy increasing network capacity

It is not given how congestion control algorithms behave in the CacheCast presence

39

CacheCast implemented in ns-2 Simulation setup:

Bottleneck link topology 100 TCP flows and

100 TFRC flows Link cache operating

on a bottleneck link

40

TCP flows consume the spare capacity TFRC flows increase end-to-end throughput

CacheCast preserves the Internet ‘fairness’

41

J. Santos and D. Wetherall, “Increasing effective link bandwidth by suppressing replicated data,” USENIX’98

A. Anand, A. Gupta, A. Akella, S. Seshan, and S. Shenker, “Packet caches on routers: the implications of universal redundant traffic elimination,” SIGCOMM’08

A. Anand, V. Sekar, and A. Akella, “SmartRe: an architecture for coordinated network-wide redundancy elimination,” SIGCOMM’09

42

I. Packet caching system can achieve near multicast bandwidth savings

II. Packet caching system requires server support

III. Packet caching system is incrementally deployable

IV. Packet caching system preserves fairness in Internet

43

I. Packet caching system can achieve near multicast bandwidth savings

II. Packet caching system requires server support

III. Packet caching system is incrementally deployable

IV. Packet caching system preserves fairness in Internet

44

I. Packet caching system can achieve near multicast bandwidth savings

II. Packet caching system requires server support

III. Packet caching system is incrementally deployable

IV. Packet caching system preserves fairness in Internet

45

I. Packet caching system can achieve near multicast bandwidth savings

II. Packet caching system requires server support

III. Packet caching system is incrementally deployable

IV.Packet caching system preserves fairness in Internet

46

Getting CacheCast into the real world

Server support

Link cache

II and III generations of router

47

48