Technology Assessment of Middleware for Telecommunications

bunchlearnedΔίκτυα και Επικοινωνίες

30 Οκτ 2013 (πριν από 3 χρόνια και 7 μήνες)

63 εμφανίσεις


Technical Information




Technology Assessment of Middleware for
Telecommunications



Feasibility Study: Application of ADSL as Transport Network



Authors:

Frank Stoinski (editor), Humboldt University Berlin/Deutsche Telekom AG

Peter Loosemore, BT


British Telecommunications
plc



Abstract

This technical information summarises the investigation of the current status of ADSL deployment
and the possible use of ADSL as the transport network access technology for stream interface
bindings. This document describes the information co
llected in this experiment about the status of
ADSL deployment in each partner’s domain, and the possible benefits of using ADSL for P910.
cu牴he牭o牥 a possible 牯admap 景爠晵牴he爠studies and subsequent expe物ments is shown.


EDIN

0121
-
0910

Project

P910

For full publication

April 2001































Disclaimer

This document contains material which is the copyright of certain EURESCOM PARTICIPANTS,
and may not be reproduced or copied without permission.

All PARTICIPANTS have agreed to full publication of this documen
t.

The commercial use of any information contained in this document may require a license from the
proprietor of that information.

Neither the PARTICIPANTS nor EURESCOM warrant that the information contained in the report
is capable of use, or that use of
the information is free from risk, and accept no liability for loss or
damage suffered by any person using this information.


EURESCOM Technical Information

page
3

(14)



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

ED䥎 0121
-
〹0
0

Abstract

The previous EURESCOM Project P715 has already built applications supporting continuous
multimedia data exchange by cont
rolling and managing audio and video streams. The underlying
networking technology for these experiments was N
-
ISDN and IP over N
-
ISDN respectively. The
current and increasingly deployed ADSL technology aims to provide the necessary capabilities to
bring t
he contents of such multimedia applications right to the end
-
user.

This experiment investigates the current status of ADSL deployment and the possible use of ADSL
as the transport network access technology for stream interface bindings. This document descr
ibes
the information collected in this experiment about the status of ADSL deployment in each partner’s
domain, and the possible benefits of using ADSL for P910. Furthermore a possible roadmap for
further studies and subsequent experiments will be shown.

page
4

(14)

EURESCOM Technical Information

EDIN 0121
-
0910



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

T
able of contents

Abstract

................................
................................
................................
................................
................................
.................

3

Table of contents

................................
................................
................................
................................
................................
.

4

1

Experiment Background
................................
................................
................................
................................
............

5

2

Questionnaire Overview
................................
................................
................................
................................
............

6

2.1

The Appointed D
ate of Availability

................................
................................
................................
...............

6

2.2

The Costs
................................
................................
................................
................................
.............................

6

2.3

Hardware to be installed in the CPE
................................
................................
................................
...............

7

2.4

Access Network and Premises Distribution Network

................................
................................
.................

7

2
.5

Access to other Transport Networks

................................
................................
................................
..............

7

2.6

Assignment of IP numbers

................................
................................
................................
...............................

8

2.7

Access Constraints
................................
................................
................................
................................
.............

8

2.8

Downstream and Upstream Transfer Rates

................................
................................
................................
...

9

2.9

Protocols

................................
................................
................................
................................
..............................

9

3

Analysis of Answers in Relation to the Project
................................
................................
................................
...

10

3.1

Stage of deployment

................................
................................
................................
................................
........

10

3.2

Costs
................................
................................
................................
................................
................................
...

10

3.3

Technical Parameter
s

................................
................................
................................
................................
......

10

3.3.1

Transfer Rates
................................
................................
................................
................................
..........

10

3.3.2

Assignment Policy for IP numbers

................................
................................
................................
......

11

3.3.3

Protocols

................................
................................
................................
................................
...................

11

3.4

Access constraints
................................
................................
................................
................................
............

11

4

Conclusions

................................
................................
................................
................................
...............................

12

4.1

Current usability
................................
................................
................................
................................
...............

12

4.2

Subsequent experiments

................................
................................
................................
................................
.

12

Appendix A

Questionnaire
................................
................................
................................
................................
...........

13

5

References
................................
................................
................................
................................
................................
..

14




EURESCOM Technical Information

page
5

(14)



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

ED䥎 0121
-
〹0
0

1

Experiment Background

Asymmetric Digital Subscriber Line (ADSL), a new modem technology, converts existing twisted
-
pair telephone lines into access paths for multimedia and high
-
speed data communications. ADSL
transmits more than 6
Mbps (optionally up to 8 Mbps) to a subscriber, and as much as 640 Kbps
(optionally up to 1 Mbps) more in both directions. Such rates expand existing access capacity by a
factor of 50 or more without new cabling. ADSL can literally transform the existing p
ublic
information network from one limited to voice, text and low resolution graphics to a powerful,
ubiquitous system capable of bringing multimedia, including full motion video, to everyone's home
this century.

ADSL will play a crucial role over the nex
t ten or more years as telephone companies enter new
markets for delivering information in video and multimedia formats. New broadband cabling will
take decades to reach all prospective subscribers. But success of these new services will depend
upon reachi
ng as many subscribers as possible during the first few years. By bringing movies,
television, video catalogues, remote CD
-
ROMs, corporate LANs, and the Internet into homes and
small businesses, ADSL will make these markets viable, and profitable, for tele
phone companies
and application suppliers alike.

This experiment aimed to get the necessary information on the current availability of ADSL for the
partners and to get a first timeline for defining experiments that use ADSL technology. Collection
of inform
ation was done using a questionnaire built up of nine questions concerning mainly the
technical part of the ADSL deployment. The questionnaire is included in the appendix of this
document.

page
6

(14)

EURESCOM Technical Information

EDIN 0121
-
0910



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

2

Questionnaire Overview

The main problem for some partners was that

information on ADSL deployment was hard to
obtain. Many of the technical parameters are hard to evaluate in a theoretical way for their benefit
to a KTN and stream data processing applications, since parameters obtained in practical tests may
differ from
the theoretically announced parameters. There follows the list of questions asked,
together with tables showing the responses from each partner.

2.1

The Appointed Date of Availability

ADSL is a modern technology that has been successfully field
-
tested by a num
ber of companies all
over the world. This question should clarify the date of availability of ADSL for every day use in a
partner’s domain. Nevertheless although there may be concrete dates for introduction of ADSL to
the public, there could be problems an
d delays raised from the introduction of the new technology.

BT

Currently on trial in London, no announcement for public availability, maybe available
for BT

DT

Available in big cities, up to 40 cities until end of 1999

NL

Field test in Amsterdam, availa
ble for KPN

NT

Available in big cities

OG

Technical trial from September 1999, Field trial from January 2000 in two major Greek
cities, commercially available middle to end 2000

TI

Currently on trial until August

2.2

The Costs

A major factor for a broad de
ployment of ADSL is the cost. Main costs arise from new hardware to
be bought, installation costs and running costs. This is particularly interesting for developing new
services based on ADSL technology and assuming that ADSL access will be a universal acc
ess
technology for most private customers.

BT

Trial costs were £30 (46.81 Euro) per month, maybe usage based billing in the future

DT

Private:

DM 99 (50.11 Euro) incl.
50 free online hours per month,

DM 149 (76.18 Euro) incl.
100 free online hours per mon
th

Business: 2.4 GByte free Data Volume

Public Internet: 1536/160: DM 2000 establishing, DM 790 per month,


2048/192: DM 2900 establishing, DM 1990 per month,


4096/384: DM 7500 establishing, DM 3290 per mo
nth,


6016/576: DM 8900 establishing, DM 4490 per month

ATM: 2048/192: DM 500 establishing, DM 390 per month,


4096/384: DM 500 establishing, DM 680 per month,


6016/
576: DM 500 establishing, DM 960 per month

NL

Not yet announced

EURESCOM Technical Information

page
7

(14)



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

ED䥎 0121
-
〹0
0



Establishing: NOK 11990 ⠱462.47 Eu牯⤬

創nning 2048/448: NOK 21850 ⠲665.21 Eu牯⤠pe爠month,

創nning 1024/384: NOK 10640 ⠱297.95 Eu牯⤠pe爠month,

創nning 640/256: NOK 8240 ⠱005.16 Eu牯
⤠pe爠month



Not yet announced, ea牬iest announcement 晩牳t hal映o映2000



Not yet announced

2.3

Hardware to be installed in the CPE

Using ADSL technology requires additional hardware to be installed in the Customer Premises
Environment. The main new har
dware is the ADSL Transmission Unit (ATU), which transmits
digital signals over ordinary copper lines. This unit is also often called an ADSL modem. When
using ADSL together with POTS, a secondary hardware, the POTS splitter, has to be installed
which sepa
rates the high frequency signals of the ATU from the low frequency signals of the
phone. In many cases additional hardware is also required to connect a computer or some other
equipment to the ATU.

BT

ADSL modem, Router, Ethernet card or ATM35
-
card, or USB

DT

ADSL modem, Router (when using more then one computer), Ethernet card

NL

ADSL modem (includes POTS splitter)

NT

ADSL modem and POTS splitter

OG

ADSL modem and POTS splitter, Router (LAN), Ethernet or ATM card

TI

ADSL modem, Ethernet card

2.4

Access N
etwork and Premises Distribution Network

What kind of services are deployable using ADSL technologies mainly depends on the architecture
of the access network and the premises distribution network. Most architectures allow for access to
the public Internet
, but for special multimedia services there is a specifically designed access
network and premises distribution network needed, which satisfies the requirements of these
services in terms of speed, delay and bandwidth. By preferring one approach over the o
ther,
companies also decide for the future whether they only want to provide Internet access or go for a
complete integration of different services into the ADSL framework.

BT

Access: Ethernet, PDN: Ethernet or ATM

DT

Access: ATM, PDN: Ethernet or ATM

NL


NT

Access: ATM, PDN: Ethernet

OG

Access: ATM, PDN: Ethernet or ATM

TI

Access: Ethernet, PDN: Ethernet

2.5

Access to other Transport Networks

The capabilities of ADSL to transmit data at rates up to 8 MBit per second to the customer allow
for the provisio
n of high
-
speed multimedia services. In most cases such multimedia services will
utilize a high
-
speed network independent from the public Internet. Such integration needs
additional hardware (access node) that concentrates data from different kinds of tran
sport networks.

page
8

(14)

EURESCOM Technical Information

EDIN 0121
-
0910



2001 EU剅S䍏M Pa牴icipants in P牯ject P910



Access to an ATM netwo牫, access to a Video se牶e爠 and 䥮te牡ctive video se牶e爬
investigation on Vi牴ual P物vate Netwo牫s



Access to an ATM netwo牫



Access to netwo牫s o晦f物ng delay
-
TV, music



Only public 䥮te牮et



Access t
o an ATM netwo牫, Vi牴ual P物vate Netwo牫s



Access to othe爠netwo牫s

2.6

Assignment of IP numbers

The way in which IP numbers are assigned is important for using ADSL technology within a KTN.
There is no problem when all participating computers use statica
lly assigned IP numbers, since the
normal IP routing mechanisms in the public Internet take care of delivering packets between the
partners. As soon as one of the involved computers has dynamically assigned IP numbers, building
up the KTN requires more ef
fort. Computers with dynamically assigned IP numbers have to
announce their currently assigned IP number every time they want to participate in the KTN. How
communication and routing in such a dynamically configured KTN is done is still an open
question, a
nd has to be solved in subsequent experiments. Another problem to be considered is the
availability of a computer with dynamically assigned IP numbers. Since a computer using
dynamically assigned IP numbers with ADSL has to do some kind of dial
-
in on start
up to get its IP
number, there seems to be no way to contact such a computer from the network in case it currently
has no assigned IP number. A further problem to be discussed is the validity of object references
for persistent objects.

BT

Currently static
ally assigned, not assigned by BT, but ISP, so maybe switching to
dynamically assigned IP numbers in the future

DT

Private: Dynamically assigned

Business: Statically assigned, 1 Class
-
C net

NL

Currently dynamically, likely to be replaced by different sch
eme

NT

Statically assigned

OG

Not yet decided

TI

Dynamically assigned

2.7

Access Constraints

The freedom to use the ADSL provided access to transport networks like public Internet whenever
it is desired is a major feature needed for building up a KTN, or t
o use ADSL for stream binding
purposes. Any access constraints limit the usability of ADSL for business networking purposes.

BT

None at present

DT

No access constraints

NL

No time constraints, field test limited to 1600 users

NT

No access constraints

O
G

No access constraints

TI

Access constraints on number of users

EURESCOM Technical Information

page
9

(14)



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

ED䥎 0121
-
〹0
0

2.8

Downstream and Upstream Transfer Rates

One of the most important factors for stream binding is the transfer rate of the underlying network.
Since ADSL is an asymmetric transmission technolo
gy, the transfer rates from a service provider to
the customer (downstream) are much higher than the transfer rates from the customer to the service
provider (upstream). When stream binding is to be used with ADSL between different partners, the
crucial as
pect is the upstream transfer rate, since it limits the transfer rate of data on the originating
partners side. When considering transfer rates to the customer from a dedicated network (like
Interactive Video Network), the downstream transfer rate is a fac
tor for evaluating the possible
performance.

BT

2 MBit/s downstream, 256 kBit/s upstream during trial test

DT

Private: 768 KBit/s downstream, ca. 352 kBits/s upstream

Business: from 1.536 MBits/s to 6016 MBit/s downstream, from 160 KBits/s to 576
KBits/s
upstream

NL

2 MBit/s downstream, 400 kBit/s upstream

NT

2 MBit/s downstream, 448 kBit/s upstream

OG`

To be decided in the first half of 2000

TI

Not yet been defined

2.9

Protocols

Another interesting aspect is the use of dedicated protocols when realizing
ADSL access.
Subsequent experiments should investigate the impact of such protocols on the usability of ADSL
for KTN and stream binding purposes, and possible limitations raised by these protocols.

BT

PPP, later maybe L2TP

DT

PPP

NL

PPP

NT

PPP

OG`

PPP

TI

PPP

page
10

(14)

EURESCOM Technical Information

EDIN 0121
-
0910



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

3

Analysis of Answers in Relation to the Project

ADSL technology seems to be in an early stage of deployment. Nevertheless there have been a
number of field tests, but regular use of ADSL technology in terms of a product is still in its
starting pha
se. Many companies are still evaluating benefits and possible marketing strategies of
ADSL. For this reasons many important factors about ADSL usability for the project are still
undefined.

3.1

Stage of deployment

ADSL access deployment and marketing has just
been started in Germany and Norway. For NT
and DT it should be possible to get a near
-
term ADSL access, although there is still some effort
needed to clarify the impact of an in
-
house telephone system on the provision of ADSL access. For
KPN ADSL access co
uld be realized even though it is still in its field test stage. For BT and TI
there are no concrete dates for public availability announced, so optimistic estimation of the time of
deployment could be up to 6 months.

3.2

Costs

The costs for ADSL access are on
ly announced for DT and NT. ADSL access for DT (at least for
the business versions) and NT seems to be a pretty expensive investment according to the
establishing and running costs. For DT there could be additional costs in the private version caused
by ti
me constraints associated with the available private ADSL products, but the costs for
additional online minutes are negligible compared to the costs for online minutes in the P715
architecture. For all other partners costs are still to be decided by the de
ploying companies.

3.3

Technical Parameters

3.3.1

Transfer Rates

There are two possible scenarios for evaluating the transfer rates:



Transmitting data between nodes with ADSL access



Transmitting data from nodes not using ADSL for access purposes to nodes with ADSL a
ccess

For the first scenario the upstream transfer rate limits the transfer rate of the whole system. As far
as the upstream transfer rates satisfies the needs of the communication between the two nodes,
ADSL access could be used within this scenario. For
the second scenario, the downstream transfer
rate realizes a fast access to a dedicated transport network and applications and services can assume
to have the full downstream transfer rate available at least from the access network termination to
the ADSL
connected node.

There are concrete upstream transfer rates from BT, DT, KPN and NT. The smallest transfer rate is
available to BT, which limits the data transfer rate in a partner
-
to
-
partner communication to 256
KBits/s. A KTN built on this technology shou
ld be fast enough and outperform the N
-
ISDN based
KTN of P715. For stream binding purposes the amount of data transported in a second should be
about 4 times higher than in the P715 configuration as far as the utilized access and transport
networks can ens
ure this transfer rate. Most of the partners have downstream transfer rates of 2
MBit/s available. This should be a good basis for accessing information from the public Internet
and for designing stream services involving service providers in the public In
ternet. Also the lowest
downstream transfer rate of 768 KBit/s for DT in the private version should be powerful enough for
new high
-
data
-
volume services. Considering the transfer rates provided with current ADSL
products, ADSL could enhance functionality a
nd performance of the KTN and of stream based
services used in P715. With the companies getting more experiences with ADSL deployment,
products using higher transfer rates could become available.

EURESCOM Technical Information

page
11

(14)



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

ED䥎 0121
-
〹0
0

3.3.2

Assignment Policy for IP numbers

The P715 KTN used staticall
y assigned IP numbers, which were routed by a router located in
Heidelberg. ADSL access is provided in two ways:



Statically assigned IP numbers



Dynamically assigned IP numbers

With statically assigned IP numbers there is no problem to built up a KTN, since

routing of IP
traffic originating by a node with ADSL access is done by the ADSL providing company or ISP.
Dynamically assigned IP numbers are a problem, since there must be a way to announce the
currently assigned IP number to other partners. Another pro
blem to be solved is the use of
persistent object references with dynamically assigned IP numbers, since at least for CORBA’s
Interoperable Object References an IP number is contained in the IIOP profile. For objects created
dynamically during one ADSL ses
sion and with a lifetime limited to this session, there is no
problem with dynamically assigned IP numbers.

3.3.3

Protocols

According to the answers on the question about additional protocols to be deployed, most partners
use PPP. Whether or not additional proto
cols have a noticeable impact on the communication is a
question to be answered in subsequent experiments.

3.4

Access constraints

Some partners have announced access constraints for the number of concurrent users. That means
that there are dependencies on othe
r customers provided with ADSL regarding the time to use
ADSL. How far this raises serious problems in the actual use of the ADSL access has to be proved
in subsequent experiments. No partner has announced any time constraints, so regardless of the
user co
nstraints; access should be available 24 hours a day.

page
12

(14)

EURESCOM Technical Information

EDIN 0121
-
0910



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

4

Conclusions

4.1

Current usability

In its current stage, the use of ADSL for project purposes is not recommended because of the very
early deployment phase. At least for DT and NT field tests have been compl
eted and ADSL access
is marketed in a regular way. For the other partners, concrete products on ADSL technology
haven’t been announced yet. Nevertheless, it is obvious that most of the partners will be able to get
ADSL access during the lifetime of this pr
oject and that the development of ADSL products
continues in a rapid way.

It is likely that the discussed ADSL parameters used in the field tests and concrete products may
change over the time in response to new demands and preferences from customers. Comp
anies and
customers are still very early in the development of ADSL products in respect to experiencing
ADSL technology. As a result, availability of appropriate ADSL products, satisfying the demands
of all partners, should be only a question of time.

4.2

Su
bsequent experiments

Since ADSL deployment promises great benefit for both customers and providers, it would be good
to have all partners keeping track of the ADSL deployment and marketing strategies in their
domain to be able to update the questionnaire o
n ADSL in September or October. Depending on
the updated results, experiments for actually getting ADSL access and testing it between partners
could be defined. At least for partners with restricted access policies to the public Internet, caused
for instan
ce through firewalls, ADSL could be an alternative for communication, and experiments
could also be defined for testing ADSL on an individual basis, i.e. building 'stand
-
alone' test
platforms with ADSL technology. For future service development it is also
important to take into
account the current transfer rates to be able to adjust the amount of data being exchanged by the
service provisioning applications to that transfer rate.

Results obtained in other experiments from Task 7 could be used to evaluate th
e benefit of ADSL
for P910 in terms of KTN and stream binding performance.

EURESCOM Technical Information

page
13

(14)



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

ED䥎 0121
-
〹0
0

Appendix A

Questionnaire

Questionnaire




What is the appointed day of availability of ADSL for the public and is ADSL
access possible for your location (restricted to special locations
[conurb
ations])?



What are the per day/month/MByte costs for ADSL?



What hardware has to be installed in the CPE?



What kind of access network and premises distribution network is used for
ADSL?



Is access provided to transport networks other than the public internet
?



When access is provided to the public internet, are IP numbers assigned
dynamically and on which policy (similar to dial
-
in, per month, per paid bill)?



Are there any access constraints (time, number of users)?



What maximal downstream and upstream transfe
r rates are promised and is
it possible to upgrade to higher rates (more money, private or business)?



What protocols are being deployed by ADSL access (PPP)?

page
14

(14)

EURESCOM Technical Information

EDIN 0121
-
0910



2001 EU剅S䍏M Pa牴icipants in P牯ject P910

5

References

[1]

ATM Forum Technical Report 001 “ADSL Forum System Reference Model”, ATM Forum

[2]

A
TM Forum Technical Report 002 “ATM over ADSL Recommendation”, ATM Forum

[3]

ATM Forum Technical Report 004 “Network Migration”, ATM Forum

[4]

ATM Forum Technical Report 005 “ADSL Network Element Management”, ATM Forum

[5]

ATM Forum Technical Report 006 “SN
MP
-
based ADSL Line MIB”, ATM Forum

[6]

ATM Forum Technical Report 007 “Interfaces and System Configuration for ADSL:
Customer Premises”, ATM Forum

[7]

ATM Forum Technical Report 010 “Requirement & Reference Models for ADSL Access
Networks: The ‘SNAG’ Docum
ent”, ATM Forum

[8]

ADSL access networks, Tom Starr and Kim Maxwell, May 1998

[9]

CCNET Internet Solutions ADSL information,
http://er.reddeer.net/cnnetinfo/adsl.htm

[10]

ADSL Forum,
http://www.adsl.com

[11]

WebProForum Tutorials, Broadband Access and Services, Asymmetric Digital Subscriber
Line, http://www.webproforum.com/agcommsys/index.html