We define a common format for the logs related to the messages dissemination. Any system that produce logs and translate them to this format will be able to use the DGS tools related to the messages dissemination (e.g. comparison between the horizons and the message journeys)

Events

The events are the following:

subscription of a node to a group (sub)
unsubscription of a node from a group (unsub)
emission of a message by a node (snd)
reception of a message by a node (rcv)

Line format

The generic line format is the following:

date event node-id [key=value]*

with:

date: the event date in ms
event: one of the 4 events detailed below
node-id: the id of the node related to the event
key=value: properties depending on the event type

the values of the arguments or properties may be in quotes

Subscription (sub) or unsubscription (unsub)

date sub|unsub node-id gid=grp-id

with:

node-id: the id of the node that (un)subscribes to a group
grp-id: the id of the group or a regex that identifies the message contents the node is interested in

Emission (snd)

date snd node-id mid=msg-id [key=value]*

with:

node-id: the id of the node that sends the message
msg-id: the id of the message

the optional properties can be:

size=size-msg: the size of the message in bytes
dst=dst-ids: the message recipients (or content description for recipients represented by a regex)
- dst-ids: dst-id[,dst-id]*
- dst-id: (node-id|grp-id)

Reception (rcv)

date rcv node-id mid=msg-id [key=value]*

with:

node-id: the id of the node that receives the message
msg-id: the id of the message

the optional properties can be:

sdr=node-id: the id of the node that forwarded the message

Top of page

Comma-Separated Values (csv)

Text file that contains tabular data.

CSV stands for Comma-Separated Values, but we consider here tab-separated values. Indeed, the CSV format is not fully standardized: CSV also denotes several delimiter-separated formats that use other field delimiters such as semicolons, tabs or spaces. The MUON parsing tools are build on the assumption that the delimiter character is a tab.

Each line of the file is a table row. Each row consists of one or more fields, separated by tabs, that form the columns of the table. Lines starting with the '#' character are considered comments and can contain the columns names (tab-separated textual values).

CSV files can be used to represent data to be displayed in graphs (plots).

Top of page

GraphStream (dgs)

DGS files are the default files of the GraphStream project. GraphStream allows the visualisation and analysis of graphs, in particular dynamic graphs through the use of its Java library.

These files allow the storage of both static and dynamic graphs using a textual yet human understandable representation, whilst maintaining an overall small file size. Using events such as the creation, removal or modification of a node or an edge, as well as the addition, removal or modification of attributes on either a node, an edge or even a graph, the dynamics of various graphs can be defined.

For more information, please consult this link.

Top of page

Horizons (hzon)

The horizons file format describes the horizons of nodes and their progression over time. The horizon of a node at some time is the set of nodes for which a journey exists starting at this node and at this time.

Format

The horizons files are composed of lines in the form

node time nrcvrs (+|-) nodes_set

where:

node is the id of the starting node
time is the start time of the journeys starting from node
nrcvrs is the number of nodes in the horizon set of node (receivers)
(+|-) indicates whether the following set has been added (+) or removed (-) from the horizon
nodes_set is the set of nodes that have been added to a removed from the horizons set of node
- nodes_set: '[' node [ ',' node ]* ']'

Example

dmis01  0 9     + [dmis05,dmis08,dmis09,dmis12,dmis13,dmis14,dmis17,dmis18,dmis20]
dmis01  1479477 8       - [dmis14]
dmis01  1491424 7       - [dmis12]
dmis01  1492648 6       - [dmis05]
dmis01  1496019 0       - [dmis08,dmis09,dmis13,dmis17,dmis18,dmis20]

Top of page

Journeys (jrney)

Introduction

A journey is a temporal directed path in a dynamic graph, i.e. a path with temporal attributes: the journey start time and the time for each edge.

the target node of a edge is the source node of the next edge in the path
the time of the first edge in the journey is greater or equal than the start time of the journey
the time of any edge is smaller or equal than the time of its following edge in the journey.

Journeys could be grouped together, thus forming a tree if:

they have the same start node and start time
for any pair of nodes (u,v), the edges from u to v have the same time in all the journeys that contain this edge A group of journeys represents a dissemination tree from a start node at a start time

A 'group of journeys' represents the dissemination of a message amongst nodes in an opportunistic network. Thus, journey files are calculated from applicative events files. A 'group of journeys' also represent 'ideal' journeys that messages could take by exploiting the contacts in an opportunistic scenario. Thus, journey files are calculated from DGS files.

Journey file format

A journey file is composed of lines, composed of tokens separated by tab characters (\t). Each journey or group of journeys is represented by several lines, described below. Groups of journeys are separated by an empty line.

File format specification:

journeys_file ::= journey_group [ '\n\n' journey_group ]*  
journey_group ::= journey_start '\n' statistics [ '\n\t' journey_edge ]*  
journey_start ::= <identifier> <start_time> <source_node>  
statistics    ::= <nb> <sum> <min> <max> <avg> <std_dev>  
journey_edge  ::= <edge_time> <depth> source_nodes target_nodes  
source_nodes  ::= nodes_list  
target_nodes  ::= nodes_list  
nodes_list    ::= '[' <node_id> [ ',' < node_id> ]* ']'

the nodes are represented by identifiers composed of any sequence of alphanumerical characters
the depth of a edge is the number of edges between the source node and the edge in the journey (the depth of the first edge is 0). If the depth is missing, its value is -1.
the time of a edge is expressed relatively to the journey start time (delay between the start time and the edge time)
the edges are sorted in chronological order
the statistics are related to the times of the edges (delays)
when source and target nodes lists are composed of more that one node, the journey edge represents several possible edges (one for each couple of source and target nodes in the lists)

Compact journey file format

In the compact journey file format are represented the journey source node and start time, and, for each edge, the time and target node. Each journey or group of journeys is represented by a single line.

journeys_file ::= journey_group [ '\n' journey_group ]*  
journey_group ::= journey_start [ journey_edge ]*  
journey_start ::= <identifier> <start_time> <source_node> <nb_nodes>  
journey_edge  ::= <target_node>,<edge_time>

the number of nodes (<nb_nodes>) does not include the source node of the journey

Top of page

The ONE (one)

The ONE is an Opportunistic Network Environment simulator. The ONE files represent the evolution of nodes connectivity in an opportunistic network.

The ONE file format is composed of lines in the form

time CONN node1 node2 up|down

where

time is the time when the event occurs
node1 and node2 are the ids of the nodes concerned by this event
up|down indicates whether an edge between the nodes is added or removed

For more information, please consult this link.

Top of page

Plot script (plot)

Text file representing the command generated and passed to Gnuplot to create a graphical representation.

For more information, please consult this link.

Top of page

Scalable Vector Graphics (svg)

Scalable Vector Graphics (SVG) is a vector image format for defining two-dimensional graphics. SVG images are defined in a vector graphics format and stored in XML text files. SVG images can thus be scaled in size without loss of quality.

SVG images are usually graphical representations of various data

Top of page

Views (view)

The views file format describes the views of nodes and their progression over time. The view of a node at some time is the set of nodes from which a journey exists arriving at this node and at this time.

Format

The horizons files are composed of lines in the form

node time nrsdrs (+|-) nodes_set

where:

node is the id of the destination node
time is the arrival time of the journeys arriving at node
nrsdrs is the number of nodes in the view set of node (senders)
(+|-) indicates whether the following set has been added (+) or removed (-) from the horizon
nodes_set is the set of nodes that have been added to a removed from the views set of node
- nodes_set: '[' node [ ',' node ]* ']'

Top of page

File types Find here all information concerning the different file types used by MUON.

Applicative events (aevt)

Events

Line format

Subscription (sub) or unsubscription (unsub)

Emission (snd)

Reception (rcv)

Comma-Separated Values (csv)

GraphStream (dgs)

Horizons (hzon)

Format

Example

Journeys (jrney)

Introduction

Journey file format

Compact journey file format

The ONE (one)

Plot script (plot)

Scalable Vector Graphics (svg)

Views (view)

Format

File types
Find here all information concerning the different file types used by MUON.