with a significant amount of biased views

Download Report

Transcript with a significant amount of biased views

Technology Trends
with a significant amount of biased views
Stefan Lindeberg
[email protected]
+46 708 842445
Who am I?
• 17 years in datacom
– IBM, NSC, Cisco
• Board member
– Xelerated, Hyglo,
Appgate, Comlase,
Lumentis, PacketFront,
Transmode
• Advisor to:
– Wavium, Startupfactory,
European Equity Partner
Infobalt 2002-10-21
Topics
• My views on
–Communications Technology
–Solutions and Products
–Business models
Infobalt 2002-10-21
17 years experience in datacom
TCP/IP ESCON
Token Ring
If there’s anything I’ve learnt:
Ethernet
Appletalk
DECNET
MPOA
KISS
OSI
X.25
3270
Keep It Simple Stupid
SNA
IPX
APPN
Bayan Wines
FDDI (It might actually work)
DLSW
Infobalt 2002-10-21
17 years experience in datacom
A good way
to avoid C&B’s
(Crash and Burn’s)
Infobalt 2002-10-21
The only thing we have learned from history is
that we don’t learn from history
Bridging Ethernet’s
• Over leased lines
• Over ATM (lane)
• Over ATM with ½
router (MPOA)
• Over IP
• Over MPLS
Infobalt 2002-10-21
Oops!
Communications
Technology
Trends
The days of the “miracle boxes”!
A telephone switch
with an integrated
Internet router, ATM
switch, windows NT
server, WWW
server and which
can act as an
firewall, load
balancer and cook
your coffee
Infobalt 2002-10-21
The days of the “miracle boxes” are over!
• Financial
– Old services becomes
more expensive to
produce but no or
limited increased
revenue
– New service becomes
as expensive as the old
to produce
• Technical
– Jack of all trades,
master of none
Infobalt 2002-10-21
Filing the dam, more broadband users then ever
Global Broadband Connections
35
30
Miljons
25
20
15
10
5
0
2001 Q4
2002 Q1
2002 Q2
2002 Q3
2002 Q4
Estimate
Infobalt 2002-10-21
64% year
over year
growth!
Next wave of growth
• Long haul networks in place
• New investments in customer access
equipment, as new customers are signed up
• Drives investments in aggregation and
metro equipment
Investments
Customer Aggregation
Infobalt 2002-10-21
Metro
Core
Aggregation and metro network technologies
As the capacity in current access networks
is exhausted a number of choices exists:
1.
Metro DWDM
•
2.
+ capacity - cost
CWDM
•
3.
+ low cost + outdoor (-) 16 wavelengths
Next generation SONET/SDH
•
4.
+ leverage installed base - cost
Optical Ethernet
•
5.
+ low cost - enterprise focused products until now
Resilient Packet Ring
•
Infobalt 2002-10-21
+ resilient – standard not ready
Technology turned into
solutions and
products
Revenue
Technology and product selection
Cost
Product
Technology
Services
Selection
Infobalt 2002-10-21
All solutions should solve a problem
• A solution in search of a problem
– A great technology, but so what?
• A solution to an old problem
– Does it still exist?
• A solution based on erroneous assumptions
– For example: humans better than computers in
doing repetitive things
Infobalt 2002-10-21
…with out adding complexity
We have enough tools and
technologies that are:
• Too complicated
• Too limited
• Too partial
• Doesn’t solve a
problem
Great way to shoot
yourself in the foot!
Infobalt 2002-10-21
Limiting factors
Human
understanding
Designing products
Coding software
Designing networks
Operating
Troubleshooting
Maintaining networks
Infobalt 2002-10-21
Getting around the problem
Bricks
Mission specific devices
– Boxes that do ONE thing
and does it well
– Boxes we can understand
“Boxed” technologies
Infobalt 2002-10-21
More narrow but yet broader
Today
Platforms and services
focus on one
application, for
example 3G
wireless.
But the same function
or service might be
needed in other
applications
Infobalt 2002-10-21
More narrow but yet broader
Example:
• Location services or “Find me”
• Needed for mobile telephony
and other applications
• Today’s system for mobile
telephony only
• Should be for all platforms:
PC, PDA’s, terminals, fixed and
mobile telephony, Digital TV
boxes etc
• Transparent “follow me”
services
Infobalt 2002-10-21
More narrow but yet broader
Example:
• Secure identification
• Needed for electronic
commerce and other
applications
• Today’s system built for
one platform, PC based ecommerce, GSM terminals
• Should be for all platforms:
PC, PDA’s, terminals, Digital
TV boxes etc
Infobalt 2002-10-21
Solutions for the future
The solutions for the future has to:
1. Solve a real problem
2. Be mission specific
– Solve a well defined problem and do it well
3. Be broader
– Solve this problem for multiple environments
i.e. Wireless, fixed, Internet and multiple
applications
Infobalt 2002-10-21
3G, 4G and so on
Perceived and marketed 3G drivers
• High bandwidth
internet and data
access
• Video conferencing
• Music
• Watch movies, video
on demand
Infobalt 2002-10-21
Mobile 3G costs per minute*
•
•
•
•
•
•
Voice 14kbps
Music MP3 200kbps
Music DTS 700kbps
Video VHS 1.5Mbps
Video DVD 5Mbps
Video DVD+ 10Mbps
$0.20
$3.00
$10.00
$22.50
$75.00
$150.00
Two 3 minute songs:
1.5h movie VHS:
*source: Östen Mäkitalo Telia
Infobalt 2002-10-21
$18
$2025
Real 3G drivers
What will drive 3G?
• Spectrum efficiency
– The available
frequencies is already
a scarce resource
• Multimedia messaging
– Greetings
– Pictures
Infobalt 2002-10-21
IP Based Mobile Terminals
Si
”The Mobile Internet
will never be larger
than a couple of
hundred meters”
Infobalt 2002-10-21
Internet
Mobile telephony is an access method to the net
• All content and all
services accessible
through internet and
TCP/IP
• TCP/IP enabled
handsets and
terminals
Infobalt 2002-10-21
Next generation wireless networks
Wireless device
with IP
Wireless router with
2.5G, 3G and 802.11x interface
Service provider
Other customer access
technologies
xDSL, CATV, Ethernet
Internet
Infobalt 2002-10-21
Location services
Security services
Policy services
Billing services
Business Models
are changing
Business Models
Still dominating: The “triple play”!
The same organization provides:
•Physical infrastructure, cables, ducts etc
•Logical infrastructure, switches, routers
•Service infrastructure, servers billing sw
Infobalt 2002-10-21
Disintegration is happening in all areas
• Virtual GSM operators
– Doesn’t own the physical network
• 3G providers build networks together
– 3G licenses and build-out expensive
• IP VPN service providers
– Reseller of Internet services
– Deliver IP VPN service
• Deregulation continuing
– New players, counties, power companies,
construction companies etc
Infobalt 2002-10-21
Broadband services business model
$
I1 I2 I3 T1 T2 V1
Telenordia, IP-only,
Tiscali, DigiSIP …
Active, IP network
Via Europa
Physical, cables
Property owners
Hammarbysjöstad
Customers
Infobalt 2002-10-21
New business model has impact on products
• New requirements on products
– Multi-operator support
– Multiple address domains
– Security
– Management
– OSS interfaces
Infobalt 2002-10-21
Conclusion:
Networking Darwinism:
Survival of the simple solutions
Simple in this context, is simple enough that humans can
understand, because humans still have to design,
maintain and operate the network
Infobalt 2002-10-21
Questions?
Comments?
Thank you!