Embed Size (px)
Citation preview
Information Resilience through User-Assisted Caching in Disruptive Content-Centric Networks Vasilis Sourlas, Leandros Tassiulas, Ioannis Psaras, George Pavlou IFIP Networking 2015
Ioannis Psaras EPSRC Fellow University College London [email protected] !
Best Paper Award
Problem Attacked
When the network gets fragmented, and given we have a number of (in-network) caches, for how long can we keep the content “alive” in caches and end-user devices?
– How do we find “alive” content (i.e., content still in caches)?
Goals
• Find ways to: – Exploit all possible sources to retrieve content when the main path
is “down” – Exploit in-network caching to prolong information lifetime in case of
disasters – Natively support P2P-like content distribution at the network layer
Starting Points
• Information-Centric Networking – Very promising future networking environment
• Information retrieval is more important than location
– Explicitly named content chunks/packets. – Request-response at the chunk/packet level. – Flexible to adaptation through its native support to caching, mobility and multicast.
• In-network opportunistic caching – Salient characteristic of ICN. – Packets are opportunistically cached in passing by nodes. – Plenty of research on the optimization in-network caching system performance.
• Disaster scenarios (earthquake, tsunami, etc.) – Usage of ICN functional parts, even when these are disconnected from the rest of
the network (IETF ICNRG working group). – Difficult in today’s networks that mandate connectivity to central entities for
communication.
ICN World
A B C
E F
D
some/weird/name
some/weird/name
ICN Routing Engine
some/weird/name
ICN: Application-layer name à Network-layer name (the network routes to the content itself by name)
Information Resilience through SIT
A B C
E F
D
R: some/weird/name
Server for content: some/weird/name
Route based on FIB
C: some/weird/name
R: some/weird/name
✗✗
Route based on SIT
C: some/weird/name
Some sh!t happened!!
Key Design challenges & Contributions
• How to augment the original NDN content router to increase information resilience under fragmentation? – How to forward Interests when network fragmented?
• What changes are required to the main ICN packets format and their processing in order to enable P2P-like content distribution?
• Can we measure information resilience? – We build Markov processes for the hit probability and
the time to absorption of an item and find lower bounds
Router Design
• Content Store (CS) • Pending Interest Table (PIT) • Forwarding Information Base (FIB)
Same to NDN original model
Satisfied Interest Table (SIT) – Keeps track of data packet next hop. – “Breadcrumbs” for user-assisted caching. – Allows a list of outgoing faces. – Similar to Persistent Interests (PI) in C.
Tsilopoulos and G. Xylomenos, “Supporting Diverse Traffic Types in ICN” ACM SIGCOMM ICN 2011.
Index
CS
Ptr Type
PIT
FIB
SIT
/a/b .
. .
. .
Content Store (CS)Name Data
/c/d 3,1
. .
/a/b 1
Satisfied Interest Table (SIT)
Name Face List
/c 0,1
. .
/a 2
Forwarding Info Base (FIB)
Prefix Face List
/c/d 2
. .
/a/b 0,3
Name Req. Faces
Face 0
Face 1
Face 2
Face 3Pending Interest Table (PIT)
Packet Processing
• Interest Packet format – Destination flag (DF) bit to distinguish whether the Interest is headed
towards content origin (DF=0), or towards neighbouring users (DF=1).
• Interest Packet processing – Normal operation (i.e., no fragmentation): Same as in NDN – Fragmentation Detected: If the Interest cannot find a match in CS, PIT and
FIB then DF is set to 1 and follows entries in SIT. – An Interest with DF=1 can be replied both by routers and by users with
matching cached content.
• Data packet processing – Exactly the same as in NDN; follow the chain of PIT entries. – A passing by Data packet installs SIT entries. – Optionally cached in CS of each passing by router (under investigation).
Performance Bounds
System model
Absorbing State Probability
Mean Time to Absorption
• Result: When the death rate of the users interested in a content item is larger than the corresponding birth rate, the item will finally get absorbed when the content origin is not reachable.
– The formula above gives us the “time to absorption”
[1] H. M. Taylor and S. Karlin, “An Introduction to Stochastic Modeling, 3rd edition”, Academic Press, 1998.
Performance Evaluation
Strategies/Policies (after the network fragmentation)
• Interest forwarding policies – SIT based forwarding policy (STB) – Flooding forwarding policy (FLD)
• Caching policies – No caching policy (NCP) – Edge caching policy (EDG) – En-route caching policy (NRT/LCE)
• Placement/Replacement policies – Least Recently Used policy (LRU)
Evaluation setup
• Tool: Icarus • Network topology: 50 nodes - Internet topology Zoo • Traffic demand: 1req/sec at each node • Request distribution: Zipf and localised, i.e., different across
different regions • Connection rate: 1 new user per sec • “Initialization period” of 1 hour. “Observation period” of 3 hours.
Network fragmentation and origin servers of all items are not reachable.
Metrics
• Satisfaction (% of issued interests). • Absorbed Items (% of content items). • Mean Absorption Time (sec). • User Responses (% of satisfied interests) • Minimum Hop Distance (hops) • Traffic overhead (hops)
Experiments
• Model validation • Impact of cache size • Impact of users’ disconnection rate.
Model Validation
Perfect match between model and simulation!
0 200 400 600 800 1000010203040506070
100002000030000
Theoretical Experimental
V=50, !=1, "=0.1, #$%=0
A
bsor
ptio
n Ti
me
(sec
)
Information Item
Impact of the cache size
Popular messages can stay in the network for hours even with modest amounts of cache.
0 5 10 15 20 25 30 35 400102030405060708090100110 V=50, !=1, "=0.1
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Satis
fact
ion
(% o
f iss
ued
inte
rest
s)
C/M (%)
0 5 10 15 20 25 30 35 400
5
10
15
20
25
30
35
40V=50, !=1, "=0.1
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Traf
fic O
verh
ead
(hops)
C/M (%)0 5 10 15 20 25 30 35 40
1,0
1,5
2,0
2,5
3,0
3,5
4,0V=50, !=1, "=0.1
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
M
inim
um H
op D
ista
nce
C/M (%)
0 5 10 15 20 25 30 35 400
1000
2000
3000
4000
5000
6000
7000V=50, !=1, "=0.1
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Mea
n A
bsor
ptio
n Ti
me
(sec
)
C/M (%)0 5 10 15 20 25 30 35 40
0102030405060708090100 V=50, !=1, "=0.1 STB-NCP
STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Abs
orbe
d Ite
ms (
% o
f ite
ms)
C/M (%)
0 5 10 15 20 25 30 35 400
10
20
30
40
509698100
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
V=50, !=1, "=0.1
Use
r Res
ps. (
% o
f res
pond
ed in
tere
sts)
C/M (%)
Impact of users’ disconnection rate
• When disconnection rate is larger than 0.2, less than 5% of the satisfied interests are served from users.
• The STB enabled mechanisms discard less popular items fast and maintain the rest items for a longer period.
0,0 0,2 0,4 0,6 0,8 1,0 1,5 2,0
18
24
30
36
80
85
STB-NCP FLD-NCP STB-EDG-LRU FLD-EDG-LRU STB-NRT-LRU FLD-NRT-LRU
V=50, !=1, C/M=5%
Satis
fact
ion
(% o
f iss
ued
inte
rest
s)
!
0,0 0,2 0,4 0,6 0,8 1,0 1,5 2,00369121518212427 V=50, !=1, C/M=5%
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Traf
fic O
verh
ead
(hops)
!0,0 0,2 0,4 0,6 0,8 1,0 1,5 2,0
1,0
1,5
2,0
2,5
3,0
3,5
4,0V=50, !=1, C/M=5%
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Min
imum
Hop
Dis
tanc
e
!
0,0 0,2 0,4 0,6 0,8 1,0 1,5 2,00102030
400
800
1200
1600
2000 V=50, !=1, C/M=5%
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Mea
n A
bsor
ptio
n Ti
me
(sec
)
!0,0 0,2 0,4 0,6 0,8 1,0 1,5 2,0
20
30
40
50
60
70
80
90
100 V=50, !=1, C/M=5%
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
Abs
orbe
d Ite
ms (
% o
f ite
ms)
!
0,0 0,2 0,4 0,6 0,8 1,0 1,5 2,00
10
20
30
40
50
STB-NCP STB-EDG-LRU STB-NRT-LRU FLD-NCP FLD-EDG-LRU FLD-NRT-LRU
V=50, !=1, C/M=5%
Use
r Res
ps. (
% o
f res
pond
ed in
tere
sts)
!
q It is very easy to make the network resilient to fragmentation
(at least in case of disasters). q The Satisfied Interest Table (SIT) is not memory-intensive –
acts like a cache. q Some (popular) content can stay in the network for hours. q Scoped flooding can improve performance significantly
(results on the way). q P2P can be supported natively in an ICN world and is very
very helpful in case of disasters/fragmentation q We’re working to incorporate the Satisfied Interest Table
(SIT) in the NDN normal operation.
Conclusions
Some Paper Highlights
• Kaito Ohsugi, Junji Takemasa, Yuki Koizumi, Toru Hasegawa, Ioannis Psaras, “Power Consumption Model of NDN-based Multicore Software Router based on Detailed Protocol Analysis”, IEEE JSAC, Series on Green Communications and Networking, 2016.
• Ioannis Psaras, Wei Koong Chai, George Pavlou, “In-Network Cache Management and Resource Allocation for Information-Centric Networks”, IEEE Transactions on Parallel and Distributed Systems (IEEE TPDS), vol. 25, issue 11, pp. 2920-2931, 2014.
• L. Saino, I. Psaras, G. Pavlou, “Icarus: a Caching Simulator for Information-Centric Networking”, Proc. of the 7th ICST SIMUTOOLS 2014, Lisbon, Portugal, March 2014
• Lorenzo Saino, Ioannis Psaras, George Pavlou, “Understanding Sharded Caching Systems”, IEEE INFOCOM 2016, to appear.
• Ioannis Psaras, Lorenzo Saino, George Pavlou, “Revisiting Resource Pooling: The Case for In-Network Resource Sharing”, in Proc. of ACM HotNets 2014, Los Angeles, California, Oct 2014.
