Scribd
Upload
20 views11 pages

Implementation of Daryn On Armstrong

The document summarizes the implementation of Daryn on the Armstrong parallel processor. It discusses how each Armstrong node models a train station and contains unique identifying information like a number and connected tracks. Nodes maintain lists to track reservation requests and currently occupying trains on their tracks. Structures are used for communication between nodes and to store train parameters. The host processor initializes the simulation by loading the executable onto each Armstrong processor. Each node executes an initialization routine to read its identity and neighboring node identities assigned by the OS. The simulation then runs until a counter reaches an end value.

Uploaded by

Dharmendra Gutta
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
20 views11 pages

Implementation of Daryn On Armstrong

The document summarizes the implementation of Daryn on the Armstrong parallel processor. It discusses how each Armstrong node models a train station and contains unique identifying information like a number and connected tracks. Nodes maintain lists to track reservation requests and currently occupying trains on their tracks. Structures are used for communication between nodes and to store train parameters. The host processor initializes the simulation by loading the executable onto each Armstrong processor. Each node executes an initialization routine to read its identity and neighboring node identities assigned by the OS. The simulation then runs until a counter reaches an end value.

Uploaded by

Dharmendra Gutta
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 11

IMPLEMENTATION OF DARYN

ON ARMSTRONG
WHAT IS ARMSTRONG

 ARMSTRONG is a loosely coupled parallel processor.

 Loosely coupled system is a group of processors which can


operate independent of each other.
IMPLEMENTATION

 The implementation consists of data structures and pseudo


codes for different functional units.

 Pseudo code is a compact and informal high-level description


of a computer programming algorithm.
Each Armstrong node models the station contains unique
information

 The node’s unique identifier number

 The number of tracks

 The protocol connections for the node

Above information is stored in a structure


FUNCTION OF NODE

Each node that control the track maintains two lists

 The first list contains the current reservation requests for that track.

 The second list stores a single entry i.e. if any train currently
occupies the track.
CONTINUE…
Two structures are used by the nodes for communication

 The first structure corresponds to the message between any two


nodes.

 The second structure is utilized by the output buffer for each


node.

 There is another structure to store the parameters of the train


computer.
HOST PROCESSOR
 The host processor outside initializes the simulation

 and loads a copy of the executable program on each of the


corresponding ARMSTRONG processors.

 In the “main” route the node reads the identifier value


propagated to it by the host processor
CONTINUE
 Each node executes the function “initial“(a subroutine
called at starting).

 This subroutine reads the identities of all nodes. It then


recalls the Armstrong OS assigned identifier for the
neighboring nodes.

 Comparison is made between nodes own identifier and


Armstrong OS assigned identifiers for neighboring
nodes.
Identifier Identifier
Value Value
a b
INITIALIZE PROTOCOL
CONNECTIONS,NODE
INFORMATION AND
ORIGINATING TRAINS

INITIALIZE THE SIMULATION

While
Simulating
=1

if
Counter=end
value

Train function read inputs,


Stop simulation track function, clear buffer,
travelling trains
THANK YOU

You might also like