Transcript The Edge of Smartness - University of Calgary

The Edge of Smartness
Carey Williamson
iCORE Chair and Professor
Department of Computer Science
University of Calgary
Email: [email protected]
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
1
Main Message
• Now, more than ever, we need “smart edge”
devices to enhance the performance,
functionality, and efficiency of the Internet
Application
Application
Transport
Transport
Network
Network
Data Link
Data Link
Physical
May 14, 2011
Core
Network
Networks Conference
Copyright © 2005 Department of Computer Science
Physical
2
Talk Outline
•
•
•
•
•
•
•
The End-to-End Principle: Revisited
The Smart Edge: Motivation and Definition
Example 1: Redundant Traffic Elimination
Example 2: TCP Incast Problem
Example 3: Speed Scaling Systems
Future Outlook and Opportunities
Questions and Discussion
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
3
The End-to-End Principle
• Central design tenet of the Internet (simple core)
• Represented in design of TCP/IP protocol stack
• Wikipedia: Whenever possible, communication
protocol operations should be defined to occur
at the end-points of a communications system
• Some good reading:
– J. Saltzer, D. Reed, and D. Clark, “End-to-End
Arguments in System Design”, ACM ToCS, 1984
– M. Blumenthal and D. Clark, “Rethinking the Design
of the Internet: The end to end arguments vs. the
brave new world”, ACM ToIT, 2001
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
4
Internet Protocol Stack
Application
Application
Application
Transport
Transport
Router
Transport
Network
Network
Network
Data Link
Data Link
Data Link
Physical
Physical
Physical
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
5
The End-to-End Principle: Revisited
• Claim: The ongoing evolution of the Internet is
blurring our notion of what an end system is
• This is true for both client side and server side
– Client: mobile phones, proxies, middleboxes, WLAN
– Server: P2P, cloud, data centers, CDNs, Hadoop
• When something breaks in the Internet protocol
stack, we have to find a suitable retrofit to make
it work properly
• We have done this repeatedly for decades,
and will likely keep doing it again and again!
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
6
(Selected) Existing Examples
•
•
•
•
•
•
•
•
•
•
•
Mobility: Mobile IP, MoM, Home/Foreign Agents
Small devices: mobile portals, content transcoding
Web traffic volume: proxy caching, CDNs
Wireless: I-TCP, Proxy TCP, Snoop TCP, cross-layer
IP address space: Network Address Translation (NAT)
Multi-homing: smart devices, cognitive networks, SDR
Big data: P2P file sharing, BT, download managers
P2P file sharing: traffic classification, traffic shapers
Security concerns: firewalls, intrusion/anomaly detection
Intermittent connectivity: delay-tolerant networks (DTN)
Deep space: inter-planetary IP
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
7
The Smart Edge
• Putting new functionality in a “smart edge”
device seems like a logical choice, for reasons
of performance, functionality, efficiency, security
• What is meant by “smart”?
– Interconnected: one or more networks; define basic
information units; awareness of location/context
– Instrumented: suitably represent user activities;
location, time, identity, and activity; perf metrics
– Intelligent: provisioning, management, adaptation;
appropriate decision-making in real-time
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
8
Example 1:
Redundant Traffic Elimination
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
9
Motivation for RTE
• A lot of the data content carried on the Internet
today is (partially) redundant
• Examples:
– Spam email that we receive (CIBC, RBC, …)
– Regular email that we receive (drafts)
– Web pages that we visit (U of X)
• It would be nice to avoid having to send this
redundant data more than once (especially on
low-bandwidth links!)
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
10
Basic Principles of RTE
• If you can “remember” what you have
sent before, then you don’t have to
send another copy
• Redundant Traffic Elimination (RTE)
• Done using a dictionary of chunks and
their associated fingerprints
• Examples:
– Joke telling by certain CS professors
– Data deduplication in storage systems (90%)
– “WAN Optimization” in networks (20%)
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
11
A Toy Example
Mary had a little lamb
Its fleece was white as snow
And everywhere that Mary went
That lamb was sure to go
It followed her to school one day
Which was against the rule
It made the children laugh and play
To see a lamb at school
May 14, 2011
Mary had a little lamb
A little pork, a little ham
Mary had a little lamb
And then she had dessert!
Networks Conference
Copyright © 2005 Department of Computer Science
12
Chunk Granularity Issue
• Object: large potential savings, but exact
hits will be very rare
• Paragraph: very few repeats though
• Sentence: some repeats, some savings
• Chunk: “just right” size and savings
• Word: lots of repeats, small savings
• Letter: finite alphabet, many hits, but
relatively high overhead to encode
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
13
Redundant Traffic Elimination (RTE)
• Purpose: Use bottleneck link more efficiently
• Basic idea: Use a cache of data chunks to avoid
transmitting identical chunks more than once
Distance
Overlap
• RTE process:
Chunk A
...
Chunk C
...
Copyright © 2005 Department of Computer Science
FP A
FP C
... ...
Networks Conference
Chunk B
...
May 14, 2011
FP B
...
• Works within and across files
• Combines caching and chunking
Chunk C
... ...
Chunk A
Chunk B
– Divide IP packet into chunks
– Select a subset of chunks
FP A = fingerprint (Chunk A)
– Store a cache of chunks at two ends
of a network link or path
– Transfer only chunks that are not cached
Chunk cache
14
Background on RTE
• Proposed by [Spring and Wetherall 2000]
–
–
–
–
Intended to augment Web caching
Proposed for IP packet level redundancy elimination
Found up to 54% redundancy in Web traffic
Applied to high-speed wired links (WAN Optimization)
• Chunking used in storage systems to avoid
storing redundant data (data deduplication)
• Can also apply this approach in WLAN context:
–
–
–
–
Increasing demand for wireless broadband
Plenty of CPU power, cheap storage available
Wireless traffic content similar to wired traffic
More efficient use of constrained wireless channel
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
15
Some References on RTE
• N. Spring and D. Wetherall, “A Protocol-Independent Technique for
Eliminating Redundant Network Traffic”, ACM SIGCOMM 2000
• A. Anand et al., “Packet Caches on Routers: The Implications of
Universal Redundant Traffic Elimination”, ACM SIGCOMM 2008
• A. Anand et al., “Redundancy in Network Traffic: Findings and
Implications”, ACM SIGMETRICS 2009
• A. Anand et al., “SmartRE: An Architecture for Coordinated Networkwide Redundancy Elimination”, ACM SIGCOMM 2009
• B. Aggarwal et al., “EndRE: An End-System Redundancy
Elimination Service for Enterprises”, USENIX NSDI 2010
• E. Halepovic et al., “DYNABYTE: A Dynamic Sampling Algorithm for
Redundant Content Detection”, IEEE ICCCN 2011
• E. Halepovic et al., “Enhancing Redundant Network Traffic
Elimination”, under review, 2011
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
16
RTE Process Pipeline
Current
 Improve traditional RTE
 Exploit traffic non-
Packet
 Packet size (bypass
technique)
 Chunk popularity (new
cache management scheme)
 Content type (content-aware
RTE)
 Up to 50% more
detected redundancy
May 14, 2011
NIC
NIC
uniformities:
Proposed
Packet
Large
enough?
No
Chunking
(no overlap)
Yes
Next
chunk
Fingerprinting
Overlap
OK?
Yes
No
Content
promising?
No
Chunk
expansion
FIFO cache
management
Forwarding
Networks Conference
Copyright © 2005 Department of Computer Science
Yes
Fingerprinting
non-FIFO cache
management
17
Forwarding
Fixed-size Chunks with Overlap
• Traditional RTE uses variable-sized chunks with expansion
– After detecting a chunk match, the matching region is expanded
– Need to store whole packets in cache
– Need full packet cache at both ends of the link
– Constrained to FIFO replacement policy
• Replaced with fixed-size chunks (64 bytes) and overlap
– Store chunks only in cache, not whole packets (less overhead)
– Full cache needed only at receiver, fingerprints only at sender
– Allows alternative cache management schemes
• Benefits of fixed-size chunks with overlap:
– Simpler technique with lower storage overhead
– Detects 9-14% more redundancy compared to 13% with
“expansion“
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
18
LSA Cache Replacement
• Frequency-based cache replacement (not FIFO)
• Exploit non-uniform chunk popularity
• Replace chunks that contributed least to savings
–
–
–
–
Track savings by chunks, not cache hits (overlap)
New metric: “total bytes saved” per chunk
LFU-like, may cause cache pollution
Need “aging” factor: purge entire cache!
• Least Savings with Aging (LSA) improves
detected redundancy by up to 12%
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
19
RTE in Wireless Traffic
• Using traces of campus WLAN traffic
• RTE applied to aggregate wireless traffic
– Savings comparable to inbound aggregate campus
traffic, but higher for outbound direction by about 30%
– Why? Inbound traffic mix similar for campus and
WLAN traffic, but differs for outbound (more P2P)
• RTE applied to individual WLAN user traffic
– 65% of users have up to 10% redundancy in traffic
– 30% of users have 10-50% redundant traffic
– 5% of users have 50% or more redundancy
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
20
Main Sources of Redundancy
Type
Value
Nulls
57.1% Consecutive null bytes
0x00000000
Text
16.7% Plain text (English)
Gnutella
HTTP
7.3%
HTTP directives
Content-Type:
Mixed
6.2%
Plain text and other chars
14pt font
Binary
5.8%
Random characters
0x27c46128
HTML
3.7%
HTML code fragments
<HTML> <p>
Char+1
3.2%
Repeated text chars
AAAAAAAz
May 14, 2011
Description
Networks Conference
Copyright © 2005 Department of Computer Science
Example
21
Content-Aware RTE
• Improvement techniques are nice,
but chunk selection is still random
• Tackle the fundamental problem of RTE:
selecting the most redundant data chunks
• Content-based vs. Random
• Exploit non-uniform content in data traffic
• RTE savings contribution by different data chunks:
– Null strings:
57%
– Text-based:
31%
– Binary, Mixed: 12%
May 14, 2011
 Select more text-based chunks
 Bypass binary data
Networks Conference
Copyright © 2005 Department of Computer Science
22
Example with 6-byte chunks (1 of 2)
Normal selection:
HTTP/1.1 200 OK<CRLF>
Server: Apache/2.2.11 (Unix)<CRLF>
Last-Modified: Mon, 25 Jan 2010 16:19:01 GMT<CRLF>
ETag: "a7046c-a6e-47dff86f24740"<CRLF>
Accept-Ranges: bytes<CRLF>
Content-Length: 2670<CRLF>
Cache-Control: maxage=3600<CRLF>
X-UA-Compatible: IE=EmulateIE7<CRLF>
Content-Type: image/png<CRLF>
Date: Fri, 29 Jan 2010 19:30:05 GMT<CRLF>
<CRLF><CRLF>‰PNG <CRLF><CRLF>IHDR Œ . ÍÊé gAMA ÖØÔOX2
tEXtSoftware Adobe ImageReadyqÉe< PLTE÷ì|''"áá®×Õ–{{iþú•
òò¼-'$þôrââ°ÆĤóóºþövîï»·«Sš•E¾¾•âÕ\±±ŽþòpKKChhYÿðmåå±YYG“’t’’‚ÿïk„„s
žžƒ<;4ôô¼ÓÔ«¢¢†uudóð¬–
“ŒDD:ëë¹33,÷÷½æ浜šŠÚÚ«\\UJE@ŠŠqÖÕ¦þ÷”»ºšÿón²²•
GECøëjŽŽy÷÷¾–
•zì쵦¦‹““zÝÝ
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
23
Example with 6-byte chunks (2 of 2)
Content-aware selection:
HTTP/1.1 200 OK <CRLF>
Server: Apache/2.2.11 (Unix) <CRLF>
Last-Modified: Mon, 25 Jan 2010 16:19:01 GMT <CRLF>
ETag: "a7046c-a6e-47dff86f24740" <CRLF>
Accept-Ranges: bytes <CRLF>
Content-Length: 2670 <CRLF>
Cache-Control: maxage=3600 <CRLF>
X-UA-Compatible: IE=EmulateIE7 <CRLF>
Content-Type: image/png <CRLF>
Date: Fri, 29 Jan 2010 19:30:05 GMT <CRLF>
<CRLF><CRLF>‰PNG <CRLF><CRLF>IHDR Œ . ÍÊé gAMA ÖØÔOX2
tEXtSoftware Adobe ImageReadyqÉe< PLTE÷ì|''"áá®×Õ–{{iþú•
òò¼-,'$þôrââ°ÆĤóóºþövîï»·«Sš•E¾¾•âÕ\±±ŽþòpKKChhYÿðmåå±YYG“’t’’‚ÿïk„„
sžžƒ<;4ôô¼ÓÔ«¢¢†uudóð¬–
“ŒDD:ëë¹33,÷÷½æ浜šŠÚÚ«\\UJE@ŠŠqÖÕ¦þ÷”»ºšÿón²²
GECøëjŽŽy÷÷¾–
•zì쵦¦‹““zÝÝ
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
24
Entropy-based bypass
Percentage of chunks
• Lower entropy means higher redundancy
• Select from chunks with entropy of 5.3 or less
• Problem: CPU time required
25%
HTML
PDF
MP3
20%
15%
10%
5%
0%
3
3.2 3.4 3.6 3.8
4
4.2 4.4 4.6 4.8
5
5.2 5.4 5.6 5.8
6
Entropy
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
25
Textiness-based bypass
• “Textiness”: proportion of plain text characters in a chunk
• Computationally simple
• Select from chunks with textiness of at least 0.9
80%
• Modest CPU demands
• Similar RTE savings
Percentage of Chunks
70%
60%
HTML
50%
MP3
40%
30%
20%
10%
0%
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Textiness value
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
26
RTE Summary
• Improves traditional RTE savings by up to 50%
• Techniques can be used individually or together
• RTE very beneficial for wireless traffic
– 30% of users have 10-50% redundant traffic
• Proposed a novel content-aware RTE
– Improve RTE savings by up to 38%
• Challenges of content-aware RTE
– Needs refinement to be able to work on real traces, or
exploit an appropriate traffic classification scheme
– Needs improvement in execution time
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
27
Example 2:
The TCP Incast Problem
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
28
Motivation
• Emerging IT paradigms
–
–
–
–
Data centers, grid computing, HPC, multi-core
Cluster-based storage systems, SAN, NAS
Large-scale data management “in the cloud”
Data manipulation via “services-oriented computing”
• Cost and efficiency advantages from IT trends,
economy of scale, specialization marketplace
• Performance advantages from parallelism
– Partition/aggregation, Hadoop, multi-core, etc.
– Think RAID at Internet scale! (1000x)
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
29
Problem Formulation
TCP retransmission timeouts
How to provide
high goodput
for data center
applications?
•
•
•
•
TCP throughput
degradation
High-speed, low-latency network (RTT ≤ 0.1 ms)
Highly-multiplexed link (e.g., 1000 flows)
Highly-synchronized flows on bottleneck link
Limited switch buffer size (e.g., 100 packets)
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
30
Some References on TCP Incast
• A. Phanishayee et al., “Measurement and Analysis of TCP
Throughput Collapse in Cluster-based Storage Systems”,
Proceedings of FAST 2008
• Y. Chen et al., “Understanding TCP Incast Throughput Collapse in
Datacenter Networks”, WREN 2009
• V. Vasudevan et al., “Safe and Effective Fine-grained TCP
Retransmissions for Datacenter Communication”, SIGCOMM 2009
• M. Alizadeh et al., “Data Center TCP”, ACM SIGCOMM 2010
• A. Shpiner et al., “A Switch-based Approach to Throughput Collapse
and Starvation in Data Centers”, IEEE IWQoS 2010
• M. Podlesny et al., “An Application-Level Solution to the TCP-incast
Problem in Data Center Networks”, IEEE IWQoS 2011
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
31
Effect of Timer Granularity
Finer granularity definitely helps a lot!
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
32
Application-layer Scheduling
Start time of the response from the i-th server:
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
33
Solution
Analytical Model
34
Copyright © 2005 Department of Computer Science
Effect of Number of Servers
• Note non-monotonic behaviour! (ceiling functions)
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
35
Summary
Summary: TCP Incast Problem
• Data centers have specific network characteristics
• TCP-incast throughput collapse problem emerges
• Solutions:
– Tweak TCP parameters for this environment
– Redesign TCP for this environment
– Rewrite applications for this environment
– Smart edge coordination for uploads/downloads
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
36
Example 3:
Speed Scaling Systems
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
37
Motivation
• Computer systems performance evaluation
research has traditionally considered throughput,
response time, delay as performance metrics
• In modern computer and communication systems,
energy consumption, dollar cost, and sustainability
are becoming more important
• Dynamic Voltage and Frequency Scaling (DVFS)
is well-supported in modern processors, but not
used particularly effectively
• Growing research interest in “CPU Speed Scaling”
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
38
Some References on Speed Scaling
• M. Weiser et al., “Scheduling for Reduced CPU Energy”, OSDI 1994
• F. Yao et al., “A Scheduling Model for Reduced CPU Energy”,
Proceedings of ACM FOCS 1995
• N. Bansal et al., “Speed Scaling to Manage Energy and
Temperature”, JACM, Vol. 54, 2007
• N. Bansal et al., “Speed Scaling with an Arbitrary Power Function”,
Proceedings of ACM-SIAM SODA 2007
• D. Snowdon et al., “Koala: A Platform for OS-level Power
Management”, Proceedings of ACM EuroSys 2009
• S. Albers, “Energy-Efficient Algorithms”, CACM, May 2010
• L. Andrew et al., “Optimality, Fairness, and Robustness in Speed
Scaling Designs”, Proceedings of ACM SIGMETRICS 2010
• A. Gandhi, “Optimality Analysis of Energy-Performance Tradeoff for
Server Farm Management”, IFIP Performance 2010
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
39
A Toy Example
• Consider 5 jobs (with
no specific deadlines)
• Scheduling policies:
– FCFS, PS, SRPT
• Simple simulator of
single-CPU system
• Plot number of active
jobs in system vs time
• Plot number of active
bytes in system vs time
Job
Arrival Size
0
1.0
5
1
2.2
2
2
2.8
3
3
3.5
1
4
4.7
4
40
Copyright © 2005 Department of Computer Science
i
Ai
Si
0
1.0
5
1
2.2
2
2
2.8
3
3
3.5
1
4
4.7
4
41
Copyright © 2005 Department of Computer Science
i
Ai
Si
0
1.0
5
1
2.2
2
2
2.8
3
3
3.5
1
4
4.7
4
42
Copyright © 2005 Department of Computer Science
No Speed Scaling
FCFS
PS
SRPT
43
Copyright © 2005 Department of Computer Science
Dynamic Speed Scaling
FCFS
PS
SRPT
44
Copyright © 2005 Department of Computer Science
Speed Scaling Summary
• A widely-applicable problem
– Small-scale: desktops, multi-core, wireless devices
– Large-scale: enterprise networks, data centers
• Mechanisms available, but policies unclear
• Interesting tradeoffs between fairness,
efficiency, cost, optimality, and robustness
• Much more work remains to be done
– Universal fairness metrics
– Worst-case bounds vs average-case performance
– Dynamic energy pricing...
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
45
Concluding Remarks
• We need “smart edge” devices to enhance the
performance, functionality, security, and
efficiency of the Internet (now more than ever!)
Application
Application
Transport
Transport
Network
Network
Data Link
Data Link
Physical
May 14, 2011
Core
Network
Networks Conference
Copyright © 2005 Department of Computer Science
Physical
46
Future Outlook and Opportunities
•
•
•
•
•
•
•
•
Traffic classification
QoS management
Load balancing
Security and privacy
Cloud computing
Virtualization everywhere
Cognitive radio networks
Smart Applications on Virtual Infrastructure
(SAVI)
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
47
For More Information
• C. Williamson, “The Edge of Smartness”,
Workshop on Data Center Performance
(DCPerf 2011), Minneapolis, MN, June 2011
• Web site: http://www.cpsc.ucalgary.ca/~carey
• Email: [email protected]
• Questions and Discussion?
May 14, 2011
Networks Conference
Copyright © 2005 Department of Computer Science
48