ECTE291 QUIZ ON LECTURE 4
Last updated 11 April 2002 10.15pm
---
establishes a new virtual circuit every time an endpoint wants to connect to
another endpoint. |
SVC
|
A
cell is a small data unit of variable size. T/False? |
F.
Of
fixed size. |
AAL
= --- and
has 4 types of data streams, which are --- |
Application
adaptation layer. CBR,
constant bit rate. VBR,
variable… Connection
oriented packet data. Connection-less
packet data. |
AAL3/4
are both for --- -oriented packet data while AAL5 is
ITU-T’s hot little number for point-to-point transmissions (in contrast to
multi-point or internetwork ones). The sublayer designed to meet the needs for AAL5 is called
--- |
Connection.
SEAL,
= simple and effective adaptation layer. |
Anyone
for alphabet soup?? Connection
between 2 end points is accomplished through TPs (= ?), VPs (= ?), and VCs (= ?). |
Transmission
paths. Virtual
paths. Virtual
circuits. Tasty,
wasn’t it? |
As
a network expands (more/less) data is carried in headers |
More (unless using ATM) |
Asynchronous
means --- in the context of TDM. |
Flexible
time slots allocated to each sending device. Not fixed. |
ATM
stands for --- And
it is --- |
Asynchronous
transfer mode A cell relay protocol that allows high-speed
interconnections. |
ATM
uses 2 types of connections. Help me out by telling me what they are. |
PVC
and SVC. And
they stand for??? |
CBR
is the right type of data stream for applications with --- delays. [ This is AAL1. ] |
Minimal. And
transmission must simulate real time. |
DEMUX
= |
De-multiplexer.
Sexy, eh? |
Easy
one this time. Name
the 3 layers defined by the ATM standard, from the top down. |
Application
adaptation layer. The
ATM layer. The
physical layer. |
If
the user access devices, called --- ---, are connected through ----,
then how are the switches connected? |
End
points. UNI
= user to network interface. NNI
= network to network interfaces. |
In
a --- ---,
data are loaded into identical cells that can be transmitted with complete
(predictability / uniformity). |
Cell
network. Both. |
In
addition to dividing the AAL by category, the ITU-T also divides it by
function. So each of the AAL categories is actually 2 layers, named --- and --- |
Convergence
sublayer (= CS) Segmentation
and reassembly sublayer (= SAR). |
In
synchronous TDM the MUX allocates (exactly the same amount / different amounts)
of time to each device – whether it’s got anything to send or not. |
Same.
|
K.E.
Batcher did something very clever to the Bannyan
switch. What? And
don’t say invented the Batcher-Bannyan switch !! |
KE
designed a switch that comes before the Bannyan
switch and sorts the incoming calls according to their final destination. To
avoid cells colliding (bang! Bang! – it’s so noisy!). The
B-B switch uses “hardware merging techniques”. |
K.E.
Batcher question 2: Another
hardware module is added between the 2 B’s. What is it? |
The
trap module is inserted, to prevent “duplicate cells” (= cells with the same
destination) from passing to the Bannyan switch
simultaneously. |
Layers.
How many (and which) layers does ATM use? |
ATM
layer, and physical layer. |
Many of the problems
associated with packet internetworking are solved by adopting --- |
Cell
networking. |
Name
2 goals that ATM achieves. And
you get a bonus pat on the back for more than 2 !! |
Fast
data transmission Reliable
– less susceptible to noise degradation Interface
with existing systems Low
cost Is
connection-oriented It
moved as many operations to hardware as possible. |
PVC
= --- |
Permanent
virtual circuit. |
So
what does asynchronous TDM do about cells that would be empty? |
Fills
them with data from other channels, that’s what. So we say that asynchronous
TDM allocated time slots “dynamically”. Memory
hook: Async is Dynamic. |
SVC
= --- |
Could
it be switched virtual circuit, I wonder. |
T/False? ATM uses asynchronous TDM. |
T. |
T/False? Differing packet lengths make traffic
unpredictable |
T,
before ATM |
T/False?
The AAL allows existing networks such
as packet networks to connect to ATM facilities. |
T. |
T/False?
The VPI is changed by the switch as a
cell goes through the switch. |
T.
The switch consults its “switching table” to know where to send it. Bonus
answer: VCI works the same way. |
TDM
= |
Time
division multiplexing. Ah
but can you explain it?? |
TDM
can be implemented in 2 ways, which are --- |
Synchronous
TDM Asynchronous
TDM |
TDM
is a digital process than can be applied when --- |
Data
rate transmission capacity of the transmission medium > data rate of the
sending and receiving devices. So there is spare capacity in the medium. So
multiple transmissions can happen at once. |
The
basic data unit in an ATM network is called a --- and it is 53 bytes long, 5
of them being allocated to --- So
what fills up the other bytes, eh? |
Cell. A
header, with VPI
and VCI – they define the transmission path for the cell from end point to
end point. |
The
data streams coming into a MUX are called --- |
Channels. MUX=
multiplexer. |
The
length of the VCI is the same when it interfaces with NNI and UNI. T/False? |
T.
both 16 bits long. |
The
length of the VPIs for UNI is --- And
for NNI is --- (Don’t
ya just love this sorta
stuff??) |
8
and 12. Just
remember VPs (vice presidents) drive V8s, the car engine. |
The
physical connection between an end point and a switch, or 2 switches, is
called --- Examples
please. |
TP (transmission path). Wire,
cable, satellite. |
The
simplest type of switch for ATM is the --- which has --- at each crosspoint. |
Crossbar. Microswitches (transistors). |
The
switch has a clock that delivers one cell to the output at each --- |
Tick.
(Easy
question! You liked that, didn’t ya?) |
Time
slots in synchronous TDM are grouped into frames, each frame consisting of
--- |
One
cycle of time slots, one or more dedicated to each sending device. But if one
is empty, bandwidth is wasted. Oh dear! We’ll have to fix that! But how? –
look for the question on Asynchronous TDM... |
To
achieve an approximation of a continuous stream of data, use what? And
why does this work? |
Cell
networking. Because
links are high speed, and cells are small. |
To
get the most out of broadband technology, traffic must be --- |
Time
division multiplexed onto shared paths. |
VBR
data is where apps generate and consume data at variable rates, within --- [ This is AAL2. ] |
established parameters. |
VC
is inside VP, inside TP. Explain. |
A
lot of virtual connections (VCs) can be bundled inside one virtual path (VP) for
transmission along one transmission path (TP). Simple as toast! (Hey, I gotta have some fun too.) |
VP
provides 1+ connection/s between --- |
2
switches. |
VPI
= ? VCI
= ? |
Virtual
path / circuit identifiers. |
What
are interleaved as they are transmitted? And why? |
Cells in cell networking. So
there are no long delays. |
What
factor is it that allows switches and terminals to treat each cell as a unit
rather than as a bit stream? |
The
predictability produced by cells being of a fixed size. Makes it cheap and
efficient. |
What’s
the name of this multistage switch with microswitches
at each stage? These microswitches route cells
based on the output port represented as a binary string. |
Bannyan. |
Which
AAL data type might get dropped if it’s not very
careful? |
AAL2,
the variable bit-rate one. |
With
crossbar switches, 2 cells can arrive at a crosspoint
at the same time. The solution to this is --- |
The
Bannyan switch. Must
be important – it had 6 slides! |
ECTE291 submenu