Section 3 of 3 - Prev - Next
All sections - 1 - 2 - 3

discarding all other cells under this circumstances is called PPD. Now if
all the cells that are the result of fragmenting a large packet will not
fit into the available buffer space (and some will be dropped) then why
continue sending only some of the cells. Just drop the entire packet (burst
of cells), which is called EPD.

So EPD acts *before* cells belonging to an AAL5 frame are admitted to the
output buffers. If a switch buffer occupancy threshold is exceeded, then
frames are discarded by EPD without even being queued in the output
buffers. On the other hand, PPD acts *after* cells of an AAL5 frame have
been admitted to a buffer. If any one cell of a particular frame is
discarded, then the rest of the cells are also discarded, since the frame
is now errored and will require retransmission anyway.

Question: PPD/EPD interaction with Traffic Policing?

Answer: One action of traffic policing is to CLP=3D1 mark (TAG) cells which
exceed a VCs specific traffic parameters. As these cells traverse an ATM
network they will be discarded IF congestion occurs at some place in the
network. Implicitly this gives CLP=3D0 (not TAGed) cells priority in that t=
he
CLP=3D1 cells will be dropped first.

It is the result of traffic policing and the operation of CLP tagging that
causes cells to be discarded, which can then trigger EPD/PPD. However it is
also possiblt for policing to be doing the right thing and, for example,
not tagging any cells, yet still output queues are congested and the need
for EPD emerges.

---------------------------------------------------------------------------
SUBJECT D26)

                   Questions about ATM addressing schemes

Question: Why are there multiple ATM addressing schemes?

Answer: According to the ATM UNI 3.x and RFC 1577, there are three
structures of ATM Address that can identify an end station.

   * 1) E.164
   * 2) NSAP
   * 3) Both

The multiple addressing schemes exist because the various companies
representing switch and service providers could not reach an agreement on
one format, split, more or less, along public network vs atm lan lines. The
way to tell what format to use is to ask your vendor (whether network
service or equipment vendor). Assumptions are risky...

During the ISDN meetings of 1984-1988 there was much discussion in ITU and
ISO regarding NSAPs and E.164. As near as I recall it came down to the idea
that E.164 does not (by itself) constitute an NSAP, but can be part of the
NSAP.

So, if you are just operating on a LAN you would use NSAP but probably not
E.164. If you are operating on an ATM network and only addressing
end-stations (and could care less about OSI) you would be OK with E.164
addressing. Finally, if you are dealing with OSI based end stations on an
ATM network you would use both, the E.164 bit gets you to the end-station
and the NSAP add-on finds the SAP at Transport Layer.

Question: Where to find info on the encoding of E.164 addresses in NSAP
address?

Answer: In general, the best place to look for answers is ISO 8348, which
is the defining standard for NSAP addresses. Annex A contains the relevant
information, section A.5.3 especially. Some information can also be found
in section 3.1.1.3 of UNI 4.0 as well.

---------------------------------------------------------------------------
SUBJECT D27)

                           What are DBR and SBR?

What are the the following Class of Services:

   * DBR - Deterministic Bit Rate
   * SBR - Statistical Bit Rate

One viewpoint.... DBR and SBR are a serious case of ITU 'Not invented
here'. DBR is a renamed CBR (Constant Bit Rate) class and SBR a renamed VBR
(Variable Bit Rate) class. Now don't ask me why the ITU did this. Granted,
the new names are perhaps 'better' in the sense that they more precisely
describe the characteristics of the class, but still..

=2E..another viewpoint... I don't think there was any 'not invented here'
involved. CBR and VBR refer to the source (cell stream) characteristics,
and DBR and SBR relate to the concept of "ATM Transfer Capabilities"
(ITU-speak) or "service categories" (ATM Forum terminology). As there is
*not* a one-to-one relationship between cell stream characteristics and the
transfer capability used to transport the cells, it would have spawned
(even more) confusion if the same names would have been used for these
different things. DBR and SBR are included in the new version of ITU I.371.
The I.371 also includes a traffic class not supported by the ATM Forum,
called ABT (Available Block Transfer).

---------------------------------------------------------------------------
SUBJECT D28)

                         What is CLP=3D0+1 all about?

The cell flow in a connection can be logically split into various cell
flows depending on the CLP value of the cell, whether it is 0 or 1.

The following are the cell flows:

   * - CLP=3D0 cell flow
   * - CLP=3D1 cell flow
   * - CLP=3D0+1 cell flow (also called aggregate cell flow)

CLP=3D0+1 cell flow is for both CLP=3D0 cells and CLP=3D1 cells. So logical=
ly, a
CLP=3D0 cell travels in 'CLP=3D0 cell flow' and 'CLP=3D0+1 cell flow' while=
 a
CLP=3D1 cell travels in 'CLP=3D1 cell flow' and 'CLP=3D0+1 cell flow'.

The connection and cell flows may be represented as follows:

        Connection
            |
            V

    ---------------------------
    ---------------     |
    CLP=3D0 Cell Flow     |
    ---------------     CLP=3D0+1 Cell Flow
    ---------------     |
    CLP=3D1 Cell Flow     |
    ---------------     |
    ---------------------------

To establish a connection we have to specify Peak Cell Rate(PCR), Sustained
Cell Rate(SCR), Maximum Burst Size(MBS) in forward and backward directions,
for each cell flow. So PCR, SCR, etc are not single values to a connection!
We must specify these values for the cell flows CLP=3D0, CLP=3D1 and CLP=3D=
0+1.
Usually CLP=3D0+1 values will be equal to or more than the sum of PCR, etc
values of CLP=3D0 and CLP=3D1 cell flows.

Depending on the type of the connection we need to specify some (not all)
values specific to some cell flows only. TM 4.0 clearly specifies which
combinations are valid (in chapter 4). For eg. Tagging can be opted only in
VBR.3 conformance defn. in which we specify values for CLP=3D0 and CLP=3D0+=
1
cell flows only.

Right now CDVT is not signalled even in UNI 4.0. Let us say it picks from a
standard table for a PCR or SCR value. The cell conformance test will be
done for every cell flow seperately. Consider a hypotheical type with
tagging option in which we must specify values of CLP=3D0 and CLP=3D0+1 cel=
l
flows only and cell conformance has to be done for PCRs of these cell
flows. A CLP=3D0 cell will be tested with GCRA(1/PCR0, CDVT0). If it is
non-conforming, the cell is deprioritized by tagging it to CLP=3D1. Now the
cell is tested with GCRA(1/PCR01, CDVT01) to check if it is conforming.
Note that at any further check point this cell will be checked only with
GCRA(1/PCR01, CDVT01) because it is no more in CLP=3D0 cell flow. A cell se=
nt
by source with CLP=3D1 is checked only with GCRA(1/PCR01, CDVT01) at any
place.

Note: PCR0 and CDVT0 are PCR and CDVT of CLP=3D0 cell flow and PCR01 and
CDVT01 are PCR and CDVT of CLP=3D0+1 cell flow.

---------------------------------------------------------------------------
SUBJECT D29)

                  Connection establishing in the ATM layer

Question: I have not been able to find information about how connections in
the ATM layer of ATM are set up. Since ATM is connection oriented the AAL
somehow must signal to the ATM layer that it wants to have a connection
open to another host. How is this signalling done?

Answer: Actually, it's not the AAL layer that originates the request for a
connection (although if one were a strict believer in network layering, one
might assume so :-). AAL just defines how information of a given type is
packaged for transporting over the ATM network. There is a signalling
protocol (which, by the way, uses AAL5) which defines a protocol which
includes the end stations, plus any relevant ATM switches along the path.

There are various entities above AAL that could determine a connection is
needed, including the LAN Emulation Client, an IP-ATM end station, a direct
video-over-ATM application, or a human network operator. If the connection
is set up via Switched Virtual Circuits (SVC's), then the protocol used is
most likely Q.2931, previously called Q93B, most commonly referenced via
the ATM Forum's specs:

   * UNI 3.0 (most commonly in use for ATM/data interoperability today),
   * UNI 3.1 (the update for Q.2931 compatibility, no functional changes)
   * UNI 4.0 (approved in 1996)

If the connection is set up by manual means, then the management interface
of your nearby switch is most relevant.

---------------------------------------------------------------------------
SUBJECT D30)

                  Information about about B-ISDN and B-ICI

B-ISUP provides the signalling requirements to support basic bearer
services and supplementary services (for Capability Set 1 and Capability
Set 2 B-ISDN) for B-ISDN applications. In the ATM scenario, the
introduction of this protocol meets the needs to support the Switched
Virtual Connections (SVCs), whereas initial ATM service supported only the
Permanent Virtual Connections (PVCs). This protocol is conceptually the
natural evolution of the ISDN User Part (ISUP) in the Broadband field, but
many important changes have been introduced:

   * the substitution of the concept of circuit (identified by the CIC)
     with that of Virtual Path/Virtual Circuit (VP/VC)
   * the substitution of the concept of connection with that of Virtual
     Path Connection (VPC)
   * a new structure of the protocol, which is now modular and, therefore,
     open for future enhancements, in terms of Supplementary Services.
   * the possibility to manage point-to-multipoint connections/calls
     (Q.2722).
   * it can manage both E.164 and AESA addresses.

B-ISUP runs over this protocol stack:

                SS7 MTP-Level 3
                Q.2140
                Q.SAAL
                ATM

and contains a specific module, called "Compatibility Process", for
managing both unrecognized signalling informations and interworking issues
with a N-ISUP (Narrowband ISUP, i.e. ISUP).

B-ICI stands for Broadband Inter Carrier Interface and is the broad term
for the interface and B-ISUP stack as described and documented by ATM
Forum.

This is a standard interface (based on the ITU-T B-ISUP) which has been
chosen by both ITU-T and ATM Forum for interconnecting *public* ATM
networks (whereas P-NNI is the standard non-SS7 non-ITU-T based interface
for interconnecting *private* ATM networks).

This protocol takes many features from ANSI B-ISUP (T1.648.1-4), especially
those needed for routing signalling messages through different vendor
networks (like the Exit Message and the Carrier Identification Code, Charge
Number, Carrier Selection Information, Outgoing Facility Identifier,
Originating Line Information parameters).

For an introduction to both ISUP and B-ISUP see

"Signalling System #7",
Travis Russel,
McGraw-Hill

For more references surf the Trillium WEB site at http://www.trillium.com

---------------------------------------------------------------------------

Section 3 of 3 - Prev - Next
All sections - 1 - 2 - 3

© allanswers.org | Terms of use