CareerCup

Table
NODE_ID Adjacent_Node_ID

This will be enough to represent a graph.

Point table
Point pair table indicates the edge in the graph

One point pair table and include each edge only once.

Am new to DB. Can anybody explain it more elaborately.

hey i could not find anything like point table as standard is it your abbreviation for any use of any data structure or any standard

Some sort of an adjacency list where node is the key which contains a multivalued attribute column neighbour.

take table as a matrix and use matrix representation of graph. suppose that a,b,c,d are node in a graph than use column name in a table as a,b,c and d. and put single row in this table corresponding to node.

take table as a matrix and use matrix representation of graph. suppose that a,b,c,d are node in a graph than use column name in a table as a,b,c and d. and put single row in this table corresponding to node.

http://bytes.com/topic/sql-server/answers/854993-speed-q-storing-large-graphs-db-inserting-several-million-rows-research-proj

Storage Engine - MEMORY.

Store Nodes in an Array

For each node, Store Its adjacencey list in a LInked List!

Store Nodes in an Array

For each node, Store Its adjacencey list in a LInked List!

Use Clustered Index method

create table having <vertex,adj.pairs> as each vertex might point to more than one vertex there is a possibility of vertex getting repeated.
So Clustered Indexing works well

For a directed graph
Vertex(ID,X,Y)
Primary Key:ID

VertexNeighbour(ID,NeighbourVertextID,EdgeVal)
Primary Key: [ID,NeighbourVertextID]
NeighbourVertextID reference ID

For a directed graph
Vertex(ID,X,Y)
Primary Key:ID

VertexNeighbour(ID,NeighbourVertextID,EdgeVal)
Primary Key: [ID,NeighbourVertextID]
NeighbourVertextID reference ID

Table Graph
( point,
x_axis,
y_axis )

data looks like

1,2,3
2,3,4
....
....

vertex(point_id, point_value) ;
edge(point_id_parent, point_id_child, edge_value);