NODE

Interfaces and Buses / Networking / NODE

Block Name: NODE

Code File Location: VisualSim/Actor/lib/NODE

Block Overview

• Used to define a Node on the network
• Can be connected to other block to construct a complete network with multiple layers of the protocol stack
• Can defined clients, servers, routers and switches to form a full system.

Description

• This block is used to define a Node within a larger network.  The routing algorithm is defined using a Routing_Table and the connections between the nodes is defined in a database.  A set of Nodes can have one Routing_Table.  A user can connect NODE blocks in a model using the connectivity definition in the  Routing Table Database.   

• This block can be used to build a network with one routing table, or several subnets each having their own routing table.  The NODE block can be designated as a Gateway node, used to link multiple subnets.   To make a Node block Gateway, select the 'Gateway_Node' parameter.  The Gateway node name must always start with "Gateway".  The Gateway block will typically have multiple Routing_Tables listed.  This is because it will reside in multiple subnets.  The Gateway node can be common to two subnets, or there can be a intermediate set of Gateway nodes to form an intermediate Gateway subnet.  Supported Gateway topologies:  [Subnet + [Gateway] + Subnet]   or   [Subnet + [Gateway] + [Gateway] + Subnet]  Each colored link represents a routing subnet domain.

• The NODE block can be used for a simple node to node topology, or it can be used in conjunction with the Layer_Protocol blocks to create Open System Interconnect (OSI) layers on top of a NODE block.  Through parameter settings in the NODE block, one can send transactions from the NODE block to the Layer_Protocol blocks via wires/ports or via virtual connections.

Flow

• The data can arrive at the Node from the Layer or from another node.  
• If the Node cannot find the Next Hop to the Destination, it sends the packet back to the Source Node.  
• If it comes from the Layers and a path exists, it updates Task_Hop and then sends it to the next Node.  
• If it came from another Node, it sends it to the Layer.  
• If this is the Destination, it immediately sends it to the Layer.
• There are two delays at the node- Data Transfer on the link (Task_Size/Bandwidth) and Propagation Delay (Distance/(Speed of Light * Propagation Constant).

   [1] The 'Routing_Table' Block provides information for the network, including "Routing_Algorithm", "Routing_Algorithm_Cost", and "Routing_Latencies".  The connectivity information can be in a Database block with the linking name matching the Routing_Table name or in a file selected in the Routing_File_Name setting.  There can be more than one Routing_Table/Database pair in a network model. 

     [2] The Node block uses a Connectionless Routing and the data structures are passed from Node block to Node block without connections or physical wires.

     [3] Routing Table Format:  

The Routing Table Database block must have the Linking_Name match the Routing_Table_Name parameter of this block.  The format of the Database entries, is shown below: 

    ID Source_Node  Destination_Node  Distance  Speed_Mbps Duplex ;       
    0   Node_1           Node_2                  10.0         100.0             true   ;
    1   Node_2           Gateway_3             10.0         100.0             true   ;      
    2   Gateway_3      Node_2                  10.0         100.0             true   ;      
    3   Node_2           Node_1                  10.0         100.0             true   ;

Note:  There must be Tabs or spaces between each of the columns.  The first row must contain the names of the columns.  Each row corresponds to one Direct connection.  There must be a space before the ;.

    All input and output links to a NODE are described separately.  This allows the user to allocate different bandwidths in each direction and to allow users to disable a single link in a single direction.  The Add and remove of link is accomplished by the NODE_Master block.  Please refer to NODE_Master block documentation for further details.

Statistics

 Users can obtain statistics for each NODE block from the NODE_Master block or through the use of a RegEx function.  The statistics output
   looks like:

   {"Source", "Destination", "Mbps", "Min_Utilization", "Mean_Utilization", "StDev_Utilization", "Max_Utilization"}

,  {"Node_1", "Node_2", "0.0", "0.0", "0.0", "0.0", "0.0"}

, {"Node_1", "Node_3", "156309.67", "0.00137", "0.00156", "1.612977E-4", "0.00319000032"}

The generated statistics are for each link and contain the Mbps and, min, mean, stdev and max of the link utilization.  The RegEx function allows one to collect all of the statistics for the entire routing table domain, using the following expression, assuming the Routing_Table name is "RT":

     getBlockStatus("RT", "stats", -1)    // Generates the statistics and resets the statistics for all  links in Routing Table- "RT"

    getBlockStatus("RT", "stats", 0)    // Generates statistics for all  links in Routing Table- "RT"

    getBlockStatus("RT", "stats", 1)    // Returns connectivity table for routing table domain

    getRoutingTableHop("RT","Node_1","Node_2")    // Returns the next node hop, if there is no hop, then return "none"

RT represents the name of the Routing Table associated with this NODE block.

Fields

There are two types of fields- one are the user input and the others are used internally.  The data structure used must contain these fields or a error will be generated.  The user fields are:

Task_Source: Message Source Node.  This is not modified by this block.

Task_Hop: Next Node in the data transfer.  The user will set it to the current node.  This block will modify the Task_Hop field based on the content of the Routing Table.  This field is modified.

Task_Destination: Final Destination Node Name.  This block uses this field to determine what the next Hop is.  This field is not updated.

Task_Size: Size in Bytes of the data, excluding Header and Trailers.  The Node block uses this field to compute the link transfer delay time. This field is not modified.

Task_ID: First Layer_Protocol block encountered will generated a sequence number.  This is a single sequence is for all Layers in all the Nodes.  When a new packets arrives a Layer for the first time, the sequence number is incremented.  For the Task_ID to be updated, the entering Task_ID=-1. This is field is not used in the Node block.  This field is not modified.

Network_Message: Currently used only for identifying Retry_, Multicast and Drop_Packet.  The Node uses this field for Retry and Drop_Packet.  If the field starts with Retry, the Node will not send this packet to the layers above.  The Node will send this packet to the next Node in the flow.  If this is the original Source Node, it will send it to the upper Layer.  If the Task_Hop cannot be found to the Task_Destination, then it sets the Network_Message to "Drop_Packet" and returns it back to the Task_Source.  If the Task_Source is the current Node, it drops the packet.  This field is modified by this block.

The Internal Fields are as follows:

Task_Retry: Number of retries for this packet.  When this number exceeds the threshold, the packet is dropped by updating Network_Message="Drop_Packet", and then returned to the top Layer_Protocol of the Source_Node.  This block ignores this fields.  This field is not modified.

Task_Time: The Node computes the latency for the Packet to transfer across the network and palces the value in this field.  This is also used to calculate utilization in the Node blocks.  This field is modified.

Task_Trace: This is a array of Node names that this packet has encountered.  The Node block appends it's name to this field.  This field is modified.

Task_Overhead: This block will maintain the amount of overhead bytes accumulated.  Subtracting this from Task_Size will give the original data size.   This is not used by this block.  This field is not modified.

Demonstrations

Network_Basic Demo Model.  This shows a basic connectionless network, the Routing Table reference and the path definition in the Database block.  

Retry Network Demo Model.  This shows a Connected Network with the layers of the OSI stack referenced.

Gateway_Connected Demo Model.  This shows the use of the Node as a Gateway.

Multi_Network_w_Intermediate_Gateway Demo Model.  This model has a intermediate of Gateway nodes.

Gateway_w_Diff_Return_Model Demo Model.  In this model, the forward and return directions use a different Gateway Node. 

Two_Networks_w_Gateways Demo Model.  This shows two Gateways between two networks.

Mulitcast Demo model showing the use of the Multicast function along with the Node block.

Parameters

Ports

Created with the Personal Edition of HelpNDoc: Full-featured EPub generator