1 Internet Performance Monitoring Update Les Cottrell & Warren Matthews – SLAC Presented at the.
The Internet: Where did it come from, what are the challenges Dr. Les Cottrell SLAC, Stanford...
-
Upload
ruth-mcbride -
Category
Documents
-
view
214 -
download
0
Transcript of The Internet: Where did it come from, what are the challenges Dr. Les Cottrell SLAC, Stanford...
The Internet: Where did it come from, what are the
challengesDr. Les Cottrell SLAC, Stanford University
School on Space Weather/GNSS-GIS-Intensive computing Basic theory and hands on experience
10-21 November 2014, Koudougou, Burkina Faso
2
Agenda
SLAC Colloquium
• Brief history of the Internet and its goals,
• Design goals
• How it has grown,
• Growth & Success
• Current challenges
• Internet NG
• What is driving the changes
1961 Vision from the Bell Phone System
IT Operations August 2012
3
You mean two business machines
thousands of mile apart actually talk to each other by
telephone
Someday there will be as much
communication between business
machines as between people
Ad from June 1961 Boys’ Life Magazine
The Start of the Internet as we know it today• 1965: Larry Roberts and Thomas Marill create the first
wide area connection via telephone lines, turns out to
be inefficient and costly.
• Kleinrock & others predicts
that packet switching
more promising
1969: the original Internet created and had 4 nodes, UCLA, Stanford Research Institute, MIT, Utah, 50kbit backbone (today scaled up a billion times)
5
Early days
IT Operations August 2012
1982: Today’s TCP/IP protocols formalized
1983: 400 nodes,
Now 750M nodes (many more hidden behind
Firewalls and home routers etc.)
1972
6
Design Goals: 1
IT Operations August 2012
• Built as a collaboration of global proportions, independent stand on own, self managed autonomous systems, decentralized (chaotic, no central control/management cf. phone system),
• Best effort, no guarantees, recovery from losses, pipelining host flow control, checksums
• non-proprietary (c.f. phone system, later network protocols: IBM’s System Network Architecture, Digital Equipment’s network, Xerox Network Services…),
7
Design Goals: 2
SLAC Colloquium
• Little focus on security • Few encryption standards until 1977
• Private/public key system not until 1976
• Needed an open system to test things
• There was little to protect
• Simple black boxes: routers (blue boxes) connect nets
do not retain detailed information about individual flows or contents,
• Packets inside envelopes, layering
independent of each other, e.g. middle layers don’t know if lower layers are wireless, satellite, copper, fibre
8
Growth in % population having Internet access
SLAC Colloquium
Factor of 1,500 growth in users in 17yrs
9IT Operations August 2012
Growth: users
– Maps from http://news.bbc.co.uk/2/hi/technology/8552410.stm
Most future user growth from developing nations
11IT Operations August 2012
Growth: Devices2013: 8.7Billion devices plugged in to Internet [Forbes]
• In 10 years factor 100 growth
• cell phones & other new classes of consumer electronics
(eBooks, tablets, Internet TV, digital picture frames …), Internet
of things,
• Even bigger is machine to machine (Internet of Things) - smart grids for energy management, smart cities, surveillance
& public safety, traffic & parking control, cars, and sensor nets
…).
12IT Operations August 2012
Growth: bandwidth
Voice
Traffic
1997 breakpoint
1970 1980 1990 2000
1Tbps
1Mbps
1Gbps
1kbps
1Pbps
Max switchSpeed
Internet Traffic, Voice traffic, Max Trunk speed, Max switch speed
2010
2000: data overtakes voice
Trunk speed doubles every 22 months, 50kps 1970 =>
100Gbps today1997: Multiplex multiple colors
over single fibre
13
International Internet Bandwidth Growth 2008-2012
Annual growth > 40% in last few years
i.e. as much capacity added in 2012 as was available in
2008
www.telegeography.com/research-services/global-bandwidth-research-service/index.html
From Telegeography
14IT Operations August 2012
Compare today with 50 years ago
If we compare the IBM Stretch supercomputer of the early 1960’s
with today’s smartphone we can see we have come a long way:
• Smartphone is much smaller, i.e. it fits in the hand versus 2500
sq. feet;
• Smartphone weighs 5 oz. versus 40,000lbs;
• Smartphone uses 10,000 times less power;
• Smartphone ~ 3000 times more compute power
• Stretch $8M, smartphone few hundred $
30 ft long
15
Supercomputing growth
http://www.top500.org/
• Top machine pass
10PFlops in 2011
• US still dominates,
followed by China, UK
and FrancePer
form
ance
10EFlops10EFlops
1GFlops
Sum
#1
#500
VendorsPerformance Share
/China/Japan
/France
17
Amazing scalability Success
• The Internet has successfully scaled from a few users
to over a billion and speed increases of seven orders
of magnitude (56kbps=>100Gbps backbone)
• From a research and education
network to a commercial network
used worldwide
However there are challenges…
18
Challenge: Internet growth
IT Operations August 2012
8 orders of magnitude growth in devices in 50 years
Population: WikipediaInternet Users: Internet World StatsDevices/user: CiscoInternet devices: Cisco
19
Challenge: demands for adddress space
SLAC Colloquium
• Internet users in developing world
• Devices per person (smart phones, tablets,
wearable devices…)
• Monitoring of equipment,
humans (e.g. medical), animals
• Machine to machine (typical
car today has 16 IP addresses)
• Smart homes, smart cities, traffic,
surveillance, safety, security, power grids …
• 99% of electronics in the world today still not
connected to the internet.
20
Challenge: Address space
SLAC Colloquium
• IPv4 address space 32 bits ~ 4 billion addresses fine for
initial usage but IANA ran out Feb 2011, APNIC Apr 2011- Recognized in 1991: By-passes evolved: private addresses (e.g.
NATs), CIDR blocks etc. - Even with that will run out in next couple of years
• Initially mainly a problem for later Internet deployment
regions (China, India, Malaysia …).
• Africa best off of regions, runs out in 2019 • Not backward compatible, not as mature as IPv4 (target
for crackers), will run both for many years so added
complexity,
22SLAC Colloquium
Challenge: MobilityComputers used to be big and did not move
As move need to change IP addresses
• Topology can change
Need persistence across links going up &
down
• Delay & disruption tolerance (e.g. for space
flights)
Mesh, sensor nets, self-organizing networks
• Bad guy may join, e.g. military position overrun,
enemy gets device, pretends to be friend
23SLAC Colloquium
Challenges: Trust – slide 1 Initial trust relationship badly broken • Not everyone has everyone else’s best interest in mind
• Organized crime, state sponsored intelligence gathering,
cyber-warfare
• Akamai observed (3Q12) attack traffic originating from 180
unique countries/regions. China top ~33%, followed by US
(13%) & Russia (5% traffic)
This is compounded with:• Naïve OS’, unpatched systems, browsers, users
• Routing mistakes (e.g. black holes)
24SLAC Colloquium
Challenges: Trust – slide 2
Freedom of information vs privacy (e.g. wikileaks)• Google/Yahoo … (has your emails), Facebook have a good
idea of who your friends are where you live, work, spend your
free time, your health, love life, political leaning
• Branching out into your realtime (Global Position System) to
give your location
• Nowhere to hide anymore
Lack of tools for strong authentication needed for Grids & cloud computingPrevalence of viruses, worms, malware, Trojan horses, Denial of Service
25SLAC Colloquium
Challenge: SPAM
Unsolicited pitches for things such as
drugs (> 60%), dating, stocks, malware
(few %) …
US leads as source of spam Sent by botnets
Networks of compromised computers, millions worldwideLocated mainly in Europe (esp East), Russia, US and India
88% of all email, 150B emails/day (Cisco)
Due to spam filters only 20% of mail received by users is spam
Huge Annual losses due to viruses for a typical 1,000-user organization will amount to over $158,000 (Radicati Group).
26SLAC Colloquium
Challenges: Capacity and net neutrality
• Does this lead to ISP
inspecting and limiting
traffic
by customer
• Leading to loss of net
neutrality
File sharing
Internet videoTV Internet video
Web data
What are the
drivers?
Can Capacity keep
up with demand
principle that Internet service providers and governments should treat all data on the Internet equally, not discriminating or charging differentially by user, content, site, platform, application, type of attached equipment, and modes of communication- Wikipedia
27
Challenge: How to change it
SLAC Colloquium
Despite the new Internet Protocol (version 6 or
IPv6) being > 15 years old, it carries < 2% traffic
How to redo a functioning production network
critical to the global economy while it continues to
run
• “Creation was completed in 6 days, but no
installed user base”
• The Internet has been smoothly taking over from
the phone network
• How does it happen next time?
28
Internet NG – slide 1
SLAC Colloquium
To borrow from John Lennon: "Imagine there's no
latency, no spam or phishing, a community of trust.
Imagine all the people, able to get online from
anywhere at any time".The goal is audacious: • To create an Internet without so many security breaches, with better
trust and built-in identity management.
• Researchers are trying to build an Internet that's more reliable, higher
performing and better able to manage exabytes of content.
• And they're hoping to build an Internet that extends connectivity to the
most remote regions of the world, perhaps to other planets.
29
Internet NG – slide 2
SLAC Colloquium
Future InterNet Design (FIND) funded by
NSF to get and implement a vision for
2020
•Launched 50 projects ($0.5-1M) in 2006, now
(2010) being narrowed down to 2-4 with up to
$9M
•Similar initiative from Europe
30
How have things changed (not your fathers Internet anymore):
Sburkina Fas , University of Koudougou, Nov 2014
Youth of today brought up with very different expectations:
• what’s a wired phone, a payphone, a modem, typewriter, encyclopedia;
• => messaging, Google searches, Multimedia Internet, video communication
(YouTube), Internet access everywhere, mobility, virtual worlds, social
networking (Facebook, Twitter), video games, shared information (anyone
can publish)
And they are tomorrow’s leaders.
In 1998 75% of all Internet users were Americans, now < 15%.
2014 global IP traffic will exceed 767 Exabytes (10^18, ¾ zettabyte)
• CAGR 34% 2009-2014
• 2014 avg monthly traffic = 32M people streaming Avatar movie in 3D
continuously for whole month
Web pages quintupled in size since 2003, objects/page increase by 14%/year,
response time bad for low bw users, for others bw kept pace
31
Internet’s Impact on development
SLAC Colloquium
• Brief history of the Internet and its goals,
• how it has grown,
• today’s major challenges,
• and future research. • The impact of the Internet on development,
• The Internet Today
• How Africa has lagged the world,
• Why does it matter?
• Is the performance for Africa improving, will it catch up?
32SLAC Colloquium
How does the Internet assist development?
• Investment in information technology plays the role of a
"facilitator" that allows other innovations to take
place. http://findarticles.com/p/articles/mi_m1093/is_3_45/ai_86517828/
• World Bank / IFC report: for every 10% increase in
high-speed Internet connections there is an
increase in economic growth of 1.3 percentage
points. http://www.infodev.org/en/Article.522.html
• Example: Uganda 15% increase in price of maize
based on improved farmer bargaining power. www.itu.int/ITU-D/.../S1-01-NG-ICT_Indicators-Tim_Kelly.pptx
• A study reported by Akamai showed that 80 new jobs are
created for every 1,000 new broadband connections
Human Development Index (HDI)
SLAC Colloquium
• HDI from United Nations (UN) International Telecommunications
Union (ITU)
• composite statistic of life expectancy, education, and
income
HDI
Th
rou
ghp
ut i
n k
ilob
its/s
ec
10,000
1,000
1010
100
Source: HDI UN/ITUThroughput SLAC PingER project
Top right = GoodNote correlationAfrica (blue) behind
The Internet: What nextDr. Les Cottrell SLAC, Stanford University
School on Space Weather/GNSS-GIS-Intensive computing Basic theory and hands on experience
10-21 November 2014, Koudougou, Burkina Faso
35
Drivers
Sburkina Fas , University of Koudougou, Nov 2014
• Social media: • 1.3B Facebook users spend on
average 21 mins/day on Facebook
• 500M tweets/day (factor 10 since 2010)
• Youtube: 1M users, 4B videos viewed/day in 2012
Video:• IPTV 11% Internet traffic in 2013• IP Video 66% of Internet traffic
20101980 20001990$0,01
$1.00
$100
$10,000
$1,000,000 Hard Disk Cost/ GByte
Komorowski
• In 2000 361M users, end 2014 3Billion
• Disk storage 1980 $300,000/Gbyte => 2014 $0.04/Gbyte, almost
factor 10Million
36
Drivers: Devices
Sburkina Fas , University of Koudougou, Nov 2014
• Mobile ,
• Machine to Machine (M2M) (Internet of Things (IOT))
Source; CiscoNumber in parentheses are % in 2013 and 2018
Global Devices
Traffic by Devices
37
Result
Sburkina Fas , University of Koudougou, Nov 2014
• Annual global IP traffic will pass the zettabyte (1,000
Exabytes=1Million Pbytes = 1Billion GBytes) threshold by
the end 2016
• IP traffic is growing fastest in the Middle East and Africa,
followed by Asia Pacific. Traffic in the Middle East and Africa will
grow at a CAGR of 38 percent between 2013 and 2018.
ExaBytes/month
140
70
0
39
More Information & questions
Sburkina Fas , University of Koudougou, Nov 2014
Video predictions from Cisco
• www.cisco.com/en/US/solutions/collateral/ns341/ns525/ns537/ns705/ns827/white_paper_c11-481360_ns827_Networking_Solutions_White_Paper.htm
l
Predictions from Network
World
• http://www.networkworld.com/news/2010/010410-outlook-vision-predictions.html?page=1
Future Internet
• www.networkworld.com/news/2010/010410-outlook-vision.html?ts
Map of Internet penetration 1998-
2010
• news.bbc.co.uk/2/hi/technology/8552410.stm
[email protected] National Accelerator Lab,2575 Sand Hill RoadMenlo Park, CA 94025
40
Africa
SLAC Colloquium
• Brief history of the Internet and its goals,
• how it has grown,
• today’s major challenges,
• and future research. • The impact of the Internet on development,
• Africa
• How Africa has lagged the world,
• Why does it matter?
• What has soccer got to do with it?
• Is the performance for Africa improving,
will it catch up?
41
Why does it matter to Africa
SLAC Colloquium
Cartograms from:www.geog.qmw.ac.uk/gbhgis/conference/cartogram.html
• African scientists isolated• Lack critical mass
Area=Internet Users 2002
• Need network to collaborate but it is terrible
Tertiary Education
So we have: Brain drain
Instead we need:• Brain gain, tap diaspora• Blend in distance learning• Provide leadership, train
trainers
Area=scientific papers