1 applications and cheetah outline escience vs. commercial networks three modes of bandwidth sharing...
TRANSCRIPT
1
Applications and Cheetah
Outline eScience vs. commercial networks Three modes of bandwidth sharing
large-m small-m, long-held calls small-m, short-duration calls
Applications
Malathi VeeraraghavanUniversity of [email protected]
MCNC meeting, April 10, 2006
2
Circuit/Virtual-Circuit technologies
eScience networks
Commercial networks
Raison d'etre
High-throughput connectivity between a few facilities
Moderate-throughput connectivity between millions of users
Key goal
QoS guarantees Bandwidth sharing(different style from TCP based sharing)
Call duration
Long-held Short-duration
Reach End-to-end Partial (HOV lanes)
3
Goal of eScience networks
On previous slide, we said "QoS guarantees"
But usage of the term "networks" implies bandwidth sharing
If bandwidth is not shared, we have a "link"
Leased line service
4
What's the difference?
leased line service service provider is given enough lead time to
add interface modules to meet request eScience networks being designed today
above is not true requests are handled by an automated
scheduler and granted or denied within a short lead time this is bandwidth sharing this is "book-ahead" or "advance reservations"
5
GMPLS control plane
How useful is it to these two types of networks?
Purpose of GMPLS control plane Bandwidth (BW) sharing (dynamic&
distributed) Provisioning (threading the circuits/VCs)
BW sharing mode Immediate request (can't specify a future call-
initiation time or call holding time in protocols) Call accepted or rejected: "call blocking"
6
Understanding call blocking mode of BW sharing
Input parameters Link capacity: e.g. 10Gbps
Express this as m channels If per-call BW is 1Gb/s, m = 10 m is comparable to the number of bank tellers
Traffic load measure of call arrival rate (calls/sec) and mean call
holding time (seconds/call)
Measure of success of BW sharing mode Call blocking probability (% of calls blocked) Link utilization
7
Strong dependence on m: if m=10, to run link at an 80% utilization level, call blocking probability will be a high 23.62%
Call blocking probability vs. m
Utilization
m is a measure of high-throughput vs. moderate-throughputdifference between eScience and commercial networksFor high-throughput, m is small
8
Dependence on call duration?(eScience: long-held; commercial: short-duration)
Consider traffic load Typically, if mean call duration (holding
time) is high, call arrival rate is low Traffic load is a product of these two
parameters For a given call blocking probability, link
utilization and m, the required traffic load is a fixed value
Dependence on call duration unclear
9
But BW sharing does depend on mean call holding time if m is small
Mean waiting time is proportional to mean call holding time Can afford to have a queueing based solution if calls are short
Large m Moderate throughput Small m
Short calls Long callsBank teller Doctor's office
High throughput
immediate-requestwith call blocking
call queueing book-ahead
10
eScience: small-m, long-held
Does a book-ahead BW sharing mechanism help with the high call blocking probability problem?
We analyzed and simulated two types of Book-Ahead (BA) mechanisms
BA-n: caller specifies n call-start time options BA-all: caller is willing to accept any open time slot
Parameters: K: reservation time horizon (in timeslots): 2000 Two call classes:
class-1 holding time: 100 (h1); 10% of calls (r1)
class-2 holding time: 300 (h2); 90% of calls (r2)
11
Call blocking probability
0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.050
0.05
0.1
0.15
0.2
0.25
0.3
0.35
Call Arrival Rate
Cal
l blo
ckin
g pr
obab
ility
PB
BA-1
Erlang-B
IR
BA-3
BA-5
BA-10
BA-all
(m=10, K=2000, l=2, h1=100, h2=300, r1=0.1, r2=0.9)
Xiangfei Zhu, [email protected]
12
Utilization
0.01 0.015 0.02 0.025 0.03 0.035 0.04 0.045 0.050.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Call Arrival Rate
Util
izat
ion
U
BA-all
BA-10BA-5
BA-3
IR & Erlang-B
BA-1
(m=10, K=2000, l=2, h1=100, h2=300, r1=0.1, r2=0.9)
Xiangfei Zhu, [email protected]
13
Observations BA-n/BA-all performs better than IR
eg., if we want to achieve 80% utilization, the call blocking probabilities using different mechanisms are IR, 22.6% BA-3, 8.3% BA-5, 4% BA-10, 2% BA-all, almost 0%
BA-1 performs worse than IR because of the interaction between the two call
classes if class-1 calls reserve timeslots with gaps, class-2
calls are denied
Xiangfei Zhu, [email protected]
14
Third type of BW sharing
Calls with small-m, short-duration Call queueing mode Have a couple of ideas, but need to study
delayed start distributed queueing at callers (like CSMA-CD)
15
Status check Outline
eScience vs. commercial networks
Three modes of bandwidth sharing large-m small-m, long-held calls small-m, short-duration calls
Applications
16
Applications (small-m, long-held)
Cheetah project supports TSI (Terascale Supernova Initiative) Very large dataset (TB) transfers Ensight for remote visualization
Other apps with high-throughput long-duration calls (good for book-ahead) Other eScience applications e-Learning (small-USBs at each desk - better quality) Access Grid, Videnet video conferencing, teleimmersion Distributed Grid computing - MPI apps, e.g. blast (recruit clusters
- 6 hours) IPTV/VOD/HDTV/Triple-play video streaming (entertainment) Asynchronous Storage Area Network support (nightly backups)
17
Applications Large-m applications
High-quality video-telephony (10Mbps) at every desk (3 min. average durations)
100MB to GB range file transfers: through web (Red Hat Linux distribution) GridFTP, PVFS, NFS/XFS web caching
Remote storage: LORS IBP Depots, synchronous SAN operations Gaming (warcraft, battlenet)
currently written for low-BW; need good Graphics cards if higher-BW is available can more information be moved with simpler
lower-cost graphics cards for a lower-lag experience? OfficeLive, "network is the computer" model, old dumb-terminal model
lower admin costs Others? Suggestions?
Small-m (high-throughput), short-duration calls 100MB to GB range file transfers Want to give 1Gbps per transfer to lower delays
18
Status check Outline
eScience vs. commercial networks
Three modes of bandwidth sharing large-m small-m, long-held calls small-m, short-duration calls
Apps for large-m video-telephony web downloads and caching
19
Video telephony Paul Sijben writes: "Today, video telephony usually implies
that a group of people gather around a table and watch a TV showing a similar group of people around another table. Personal video telephony usually means watching postage stamp sized people in a PC screen, whose image is refreshed occasionally."
Also, "As has been known for quite some time, the nuances of face, body and arm gestures add a wealth of information to communication."
Can we exploit higher-speeds (10Mbps, OC1, OC3 rates) for better TV-like video-telephony? Delay requirement: 150ms one-way Compress less, use more bandwidth for lower latency and
higher quality
20
Sony SNC-RZ30 Camera • Ethernet output:
• 640 x 480 @ 30 FPS
• ~ 8Mb/s using Motion-JPEG
• Built-in web server
• Total system latency
• ~109ms
(90-160ms observed)
• Timer courtesy of www.Auvidea.com
Tyson Baldridge, [email protected]
21
Video Products Group – VPG5720 Maps video signal onto a Sonet OC3 I/O Connections:
Video: SDI or NTSC/PAL Audio: AES/EBU or analog
A/V Sync within 10mS Unidirectional: need a TX & RX module at both
endpoints. Need one chassis per endpoint: VPG6200 Esoteric solution, but ideal for testing best-case
performance Any experience?
Tyson Baldridge, [email protected]
22
Other issues Automating "camera-man," "director" roles Movement sensor based positioning of camera? Lighting issues Eye contact Placement in offices LCD projectors on to walls? Multiple camera feeds - hence the director role? Not really teleimmersion but a better experience to
improve remote communications (save energy costs, less travel!)
Suggestions?
23
Status check Outline
eScience vs. commercial networks
Three modes of bandwidth sharing large-m small-m, long-held calls small-m, short-duration calls
Apps for large-m video-telephony web downloads and caching
24
File transfer application
Two CHEETAH solutions Web file transfers across CHEETAH (end-to-end) Web caching (partial-path circuits)
25
Web File Transfer on CHEETAHConsists of a software package, called WebFT Leverages CGI for deployment without
modifying web client and web server software
Integrated with CHEETAH end-host software APIs to provide use of CHEETAH network in a mode transparent to users
Xiuduan Fang, [email protected]
26
WebFT Architecture
Web serverWeb client
Web Server (e.g. Apache)
CGI scripts (download.cgi &
redirection.cgi
URLResponse
WebFT sender
OCS API RD API
RSVP-TE API
C-TCP API
Web Browser(e.g. Mozilla)
WebFT receiver
RSVP-TE API
C-TCP API
Control messages via Internet
Data transfers via a circuit
OCS daemon
RD daemon
RSVP-TE daemon
RSVP-TE daemon
Xiuduan Fang, [email protected]
Cheetah end-host software APIsand daemons
Cheetah end-host software APIsand daemons
OCS: Optical connectivity serviceRD: Routing DecisionC-TCP: Circuit TCP
27
Experimental Testbed and Results for WebFT--cont.
Zelda3 and Wukong: Dell machines, running Linux FC3 and ext2/3, with RAID-0 SCCI disks
RTT between them: 24.7ms on the Internet path, and 8.6ms for the CHEETAH circuit.
CHEETAH Network
CHEETAH Network
InternetInternet
zelda3
NIC I
NIC II
wukong
NIC I
NIC II
IP routers IP routers
Sycamore SN16000MCNC, NC
Sycamore SN16000Atlanta, GA
Xiuduan Fang, [email protected]
28
Experimental Results for WebFT--cont.
The web page to test WebFT
Test parameters: Test.rm: 1.6 GB, circuit rate: 1 Gbps
Test results throughput: 680 Mbps, delay: 19 s
Xiuduan Fang, [email protected]
29
Web caching (partial-path circuits)
Uses the web proxy software package, squid, to make CHEETAH accessible to non-CHEETAH hosts.
Improves web performance by Breaking up the long-distance connectionless
(CL) path into a partial circuit through CHEETAH, and two low-RTT CL sub-paths
Leverages web caching protocols
Xiuduan Fang, [email protected]
30
The Framework for using web caching and partial-path circuits
InternetInternet
CHEETAHCHEETAH
CHEETAHApplication
Gateway (CAG)
CHEETAHApplication
Gateway (CAG)
Web client Web server
Original HTTP messages HTTP
messages
HTTP and ICP messages
HTTP messages
squidsquid
Xiuduan Fang, [email protected]
31
Experimental Testbed
wuneng at MCNCNIC I: 128.109.34.22
NIC II: 152.48.249.103
zelda1 at AtlantaNIC I: 130.207.252.131
NIC II: 10.0.0.11
Web client
CHEETAH CAG CAG
Web serverBallstein.cs.virginia.edu
Configure caching hierarchy such that zelda1’s NIC II is wuneng’s parent.
Configure CAGs to cache file with the size < 4 GB RTT for the Internet path:
ballstein and wuneng: 14.6 ms RTT for the CHEETAH path
wuneng and zelda1: 8.9 ms
Xiuduan Fang, [email protected]
32
Experimental Results for Web Partial CO Transfer
Web server parameters total RTT (ms) through
CHEETAH
file size
(MB
)
delay (s) through
name location
RTT (ms) with CHEETAH, cached?
Internet
path
zelda1
ballstein
No Yes
www.kernel.org
Ashland,
Oregon
61.6 86.0 14.6 + 8.9 + 61.6 =
85.1
48 33 18 70
internap.dl.sourceforge.net
Atlanta, GA
6.0 32.0 14.6 + 8.9 + 6.0 =
29.5
113 140 36 520
www.cs.virginia.edu
Charlottesville
, VA
25.0 <1 14.6 + 8.9 + 25.0 =
48.5
113 50 30 14
Xiuduan Fang, [email protected]
33
Future Work
Decide whether to use a partial-path circuit by examining the client’s geographic location relative to server
Use squid to connect multiple circuit/VC networks to further reduce RTT
Test the scalability and reliability of squid
Model, analyze and measure performance improvement by using a partial-path circuit
Xiuduan Fang, [email protected]
34
Thanks for listening!
Conclusions: Add MPLS and/or SONET to create large-m
(moderate-BW) mode of sharing Enable partial-path circuits - through Policy Based
Routing to isolate flows (trigger signaling from host) Avoiding BW partitioning for IR and BA is hard
because IR calls have unlimited holding time while BA needs specified holding time Centralized scheduler per domain is acceptable if
BA load is low
Suggestions, comments, thoughts? email address: [email protected]
35
CentuarFastIron
FESX448
1GCompute-0-4 152.48.249.6
Orbitty Compute Nodes
1GOC192 OC192 GbE
1-8-331-8-34
1-8-35
1-8-36
1-6-1
1-6-171-8-37
MCNCCatalyst
7600
H
H
H
H
H1G1G
1G
1G
1-7-1
Compute-0-3 152.48.249.5
Compute-0-2 152.48.249.4
Compute-0-1 152.48.249.3
Compute-0-0 152.48.249.2
1G
1G1G
1G
Wukong 152.48.249.102
1-8-381-7-17
cheetah-nc
3x1G VLAN
OC192
1-6-1
1-6-17
10GbE
1-7-1
GbE
1-7-33
1-7-34
1-7-35
1-7-36
1-7-37
1-7-38
1-7-39
1GZelda1 10.0.0.11
H
H
H1G
1GZelda2 10.0.0.12
Zelda3 10.0.0.13
1G
1G
Zelda4 10.0.0.14
H
H
Zelda5 10.0.0.15
2x1G MPLS tunnels
1G1G
Cheetah-atl
OC-192 lamda
10GbEGbE
1-7-33
1-7-34
1-7-35
1-7-36
Cheetah-ornl
1-7-1 1-6-1
OC192
X1(E)UCNS1GFC1G
1G
JuniperT320
JuniperT320
1G
1G
Force10E300
switch
ORNL
Atlanta
NC
Direct fibers
VLANs
MPLS tunnels
Wuneng 152.48.249.103H1-8-39
H1G
UVa Catalyst
4948
WASHHOPI
Force10
WASHAbileneT640
NCSUM20
2x1G MPLS tunnels
CUNYFoundry
NYCHOPI
Force10
1G
1GUVa host H
CUNY host
H
1GUVa
CUNY
CHEETAH Network
Xuan Zheng, [email protected]
36
Current status
Thanks to HOPI: UVA and CUNY connectivity to CHEETAH almost done
Software completed RSVP-TE end host clients C-TCP transport protocol OCS - DNS based solution to check connectivity C-VLSR for Ethernet switches WebFT application
Current focus: TSI support + growth of network + large-m
applications