Transcript 0810session-bulkQoS

session QoS vs bulk QoS
Bob Briscoe
Chief Researcher, BT Group
Oct 2008
Q. bulk or session QoS?
A. bulk, but BE bulk QoS
QoS bypass
End users
ASPs
ISPs
as ASPs
endpoint transport resource sharing
€
€€€€/b
ISPs
peak access rate
vol cap DPI
TCA
Diffserv
FIFO
IP
flowspec
edge
AC
local
e2e
AC
policing
resource
sharing /
scheduling
link technologies
• QoS = differentiated congestion delay & bandwidth
• as link rates increase, congestion delay becoming a non-problem
• all the bandwidth-demanding applications are taking the QoS they need
•
2
just taking a larger than average cost share of the best efforts service
the information supermarket
self check-out
aisles
3
check-outs
QoS interconnect
ISPs
as ASPs
self check-out
check-outs
self check-out
check-outs
self check-out
crosssubsidy
aisles
aisles
• evolution by company death is too slow
• years
• need market evolution (by financial perf)
• months or weeks
4
aisles
check-outs
QoS interconnection includes BE QoS
• QoS interconnection is not just about explicit QoS mechanisms
• starts with visibility of BE costs
• including at interconnect [Laskowski06, Briscoe05]...
State of the Internet
Future of the Internet
45.3%
45.3%
40.0%
40.0%
27.8%
27.8%
20.0%
15.7%
15.0%
20.0%
15.0%
7.5%
5.0%
Hogs
Pow er Users
Up & Comers
% of Total Subscribers
Middle Children
% of Traffic
3.8%
Barely Users
% of Cost
20.0%
15.7%
20.0%
7.5%
5.0%
Hogs
Superpower
Users
Pow er Users
Up & Comers
% of Total Subscribers
Middle Children
% of Traffic
• this is how to get to this future
• where apps minimise cost, even if they transfer large volumes
• (limiting peak volume will wrongly cap BitTorrent DNA)
5
% of Cost
3.8%
Barely Users
automatic interconnect
usage cost allocation
sender marks 3%
of packets
legend:
re-ECN
downstream
congestion
marking [%]
lightly congested link
marking 0.2%
of packets
NA
highly congested link
marking 2.8%
of packets
marking in 2.8%
of packets crossing
interconnect
NB
ND
receiver
6
NC
interconnect aggregation
legend:
re-ECN
downstream
congestion
marking [%]
simple internalisation of all externalities
'routing money'
area =
instantaneous
downstream
congestion
volume
NA
€
solution
0|0|2|7|6|0|5
just two counters at border,
one for each direction
NB
ND
meter monthly bulk volume
of packet markings
= aggregate downstream
congestion volume in flows
without 7measuring flows
bit rate
NC
differentiated services
& admission control
Acceptable Use Policy
Your 'congestion volume' allowance:
1GB/month (= 3kb/s continuous)
This only limits the traffic you can try to
transfer above the maximum the
Internet can take when it is congested.
Under typical conditions this will allow
you to transfer about 70GB per day.
If you use software that seeks out
uncongested times and routes, you will
be able to transfer a lot more.
just happen
• as an attribute of the customer
contract, not the network
• customer can roam without
changing network
Your bit-rate is otherwise unlimited
bulk
congestion
policer
Internet
0%
0|0|2|7|6|0|5
2 Mb/s
0.3Mb/s
6 Mb/s
•
•
operators can synthesise a carrier
grade admission control service
out of this
see pre-congestion notification
8 working group at the IETF
(PCN)
0.1%
0.3%
congestion
summary
• everyone's got their eye on the wrong balls
• volume

cost (congestion)
• AF, EF & session QoS

BE QoS cost policing
• intra-domain & inter-domain
9
session QoS vs bulk QoS
Q&A
refs
spare slides
more info
interconnected visibility of BE cost
•
The Internet's missing link: rest of path metrics at interconnect
[Laskowski06] Paul Laskowski and John Chuang, "Network Monitors and Contracting Systems: Competition and Innovation" In: Proc. ACM
SIGCOMM'06, Computer Communication Review 36 (4) pp. 183--194 (September, 2006)
•
A way to do rest of path metrics
[Briscoe05] Bob Briscoe, Arnaud Jacquet, Carla Di-Cairano Gilfedder, Andrea Soppera and Martin Koyabe, "Policing Congestion Response in
an Inter-Network Using Re-Feedback“ In: Proc. ACM SIGCOMM'05, Computer Communication Review 35 (4) (September, 2005)
pre-congestion notification (PCN)
•
Diffserv’s scaling problem
[Reid05] Andy B. Reid, Economics and scalability of QoS solutions, BT Technology Journal, 23(2) 97–117 (Apr’05)
•
PCN interconnection for commercial and technical audiences:
[Briscoe05] Bob Briscoe and Steve Rudkin, Commercial Models for IP Quality of Service Interconnect, in BTTJ Special Edition on IP Quality of
Service, 23(2) 171–195 (Apr’05) <www.cs.ucl.ac.uk/staff/B.Briscoe/pubs.html#ixqos>
•
IETF PCN working group documents
<tools.ietf.org/wg/pcn/> in particular:
[PCN] Phil Eardley (Ed), Pre-Congestion Notification Architecture, Internet Draft <www.ietf.org/internet-drafts/draft-ietf-pcn-architecture-06.txt>
(Sep’08)
[re-PCN] Bob Briscoe, Emulating Border Flow Policing using Re-PCN on Bulk Data, Internet Draft
<www.cs.ucl.ac.uk/staff/B.Briscoe/pubs.html#repcn> (Sep’08)
these slides
<www.cs.ucl.ac.uk/staff/B.Briscoe/present.html>
11
shouldn't network charge more
for lower congestion?
• apologies for my sleight of hand
•
•
•
actually aiming to avoid congestion impairment (loss / delay)
congestion marking = congestion avoidance marking
alternatively, congestion marking = price marking
• clearly should charge more for higher 'price marking'
• Diffserv example may help [Gibbens02]
marking
probability
pg
qg+qb
strict priority (g)
best efforts (b)
weak competition
price of expectation of better service
12
arbitrarily
higher pg >> pb
perfect competition pb
price differential  cost differential
pg  pb
qb
IP routers
Reservation
enabled
RSVP/PCN
gateway
PCN &
Diffserv EF
Data path processing
PCN system arrangement
1
Reserved flow processing
2
Policing flow entry to P
4
Meter congestion per peer
3
Bulk pre-congestion marking
P scheduled over N
highlighting 2 flows
table of
PCN fraction
per aggregate
(per previous
big hop)
SIP
call
server
RSVP/RACF per flow
reservation signalling
1
2
expedited forwarding,
PCN-capable traffic
(P)
3
3
3
(P)
13
b/w
mgr
(P)
non-assured QoS
(N)
3
4
1
reserved
PCN
marking
probability of
PCN
packets
virtual queue
(bulk token bucket)
Prob
1
Pre-Congestion Notification
(algorithm for PCN-marking)
X = configured
admission control capacity
for PCN traffic
X ( < 1)
Yes
PCN packet queue
P
PCN pkt?
No
•
1
Expedited
Forwarding
2
3 3
3 3
4
Non-PCN packet queue(s)
N
virtual queue (a conceptual queue – actually a simple counter):
–
drained somewhat slower than the rate configured for adm ctrl of PCN traffic
–
therefore build up of virtual queue is ‘early warning’ that the amount of PCN traffic is
getting close to the configured capacity
–
NB mean
number of packets in real PCN queue is still very small
14
1
the Internet
value-based charges
over low cost floor
• over IP, currently choice between
designed for competitive pressure
towards true marginal cost
A. “good enough” service with no QoS costs (e.g. VoIP)
– but can brown-out during peak demand or anomalies
B. fairly costly QoS mechanisms – either admission control or generous sizing
• this talk: where the premium end of the market (B) is headed
•
a new IETF technology: pre-congestion notification (PCN)
•
service of ‘B’ but mechanism cost competes with ‘A’
–
assured bandwidth & latency + PSTN-equivalent call admission probability
–
fail-safe fast recovery from even multiple disasters
• core networks could soon fully guarantee sessions without touching sessions
•
some may forego falling session-value margins to compete on cost
app signal (SIP) per session
QoS admission
priority forwarding bulk data
& PCN
S
NA
15
NB
R
ND