Danfoss VLT 8000 AQUA Operations Instructions download pdf (Page 9)

8

MG.60.N2.02 - VLT is a registered Danfoss trademark

VLT

®

5000 / VLT

®

6000 HVAC / VLT

®

8000 AQUA

Collision

Detection

Network

Management

Depending on the level of a given application,

a LonWorks network may or may not require

the use of a network management node. A

network management node performs

management functions, such as:

• Find unconfigured nodes and

download their network addresses.

• Stop, start, and reset node

applications.

• Access node communication statistics.

• Configure routers and bridges.

• Download new applications programs.

• Extract the topology of a running

network.

The LonTalk protocol uses a unique collision

avoidance algorithm which allows an

overloaded channel to carry near to its

maximum capacity, rather than reducing its

throughput due to excessive collisions

between messages. When using a

communications medium that supports

collision detection, such as twisted pair, the

LonTalk protocol can optionally cancel

transmission of a packet as soon as a collision

is detected by the transceiver. This option

allows the node to immediately retransmit any

packet that has been damaged by a collision.

Without collision detection, the node would

wait the duration of the retry time to notice

that no acknowledgment was received. At that

time it would retransmit the packet, assuming

acknowledge or request/response service.

For unacknowledged service, an undetected

collision means that the packet is not received

and no retry is attempted.

knowledgments are generated by the network

host processor without intervention of the ap-

plication. Transaction IDs are used to keep

track of messages and acknowledgments so

that the application does not receive dupli-

cate messages.

An equally reliable service is

request/response

,

where a message is sent to a node or group

of nodes and individual responses are ex-

pected from each receiver. Incoming

messages are processed by the application

on the receiving side before a response is

generated. The same retry and time-out op-

tions are available as with

acknowledged

service. Responses may include data, so that

this service is particularly suitable for remote

procedure call or client/server applications.

Next in reliability is

unacknowledged repeated

.

Messages are sent multiple times to a node

or a group of nodes with no response

expected. This service is typically used when

broadcasting to large groups of nodes when

traffic generated by all the responses would

overload the network.

The final method in reliability is

unac-

knowledged

, where a message is sent once

to a node or group of nodes and no response

is expected. This option is typically used when

the highest performance is required, network

bandwidth is limited, and the application is

not sensitive to the loss of a message.

Message

Passing

(continued)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 ... 43 44

Comments to this Manuals

No comments