Effects of Buffer Size on Delay Tolerant Routing

In this post, we look at how buffer size affects, if at all, the performance of the routing protocols in DTNs. For this purpose, we will consider the following five routing protocols:

• Epidemic
• PROPHET
• Spray-and-Wait (SnW)
• First Contact (FC)
• Direct Delivery (DD)

 Detailed discussion of these protocols is scoped out here. We just note that in case of Epidemic, there is unlimited replication of the messages. In PROPHET, however, the replication is usually less than that of Epidemic. On the other hand, SnW has a fixed limit (L) on possible number of replications of a message. Finally, FC and DD involve message forwarding -- not replication. So, in the latter cases, there is always a single copy of any message in the DTN.

We will consider the buffer sizes from 20 MB to 180 MB, both inclusive, in steps of 20 MB so that we have total 9 different buffer sizes. We will use the real-life connection traces from Infocom'06. Therefore, we will need to simulate 5 * 9 = 45 scenarios to get the relevant results. The contents of the default_settings.txt file are as shown below.

#
# Different routing protocols with 78 nodes using connection traces from
# Infocom'06.
#
# 
# Group.router: 5
# Group2.bufferSize: 9
# Total scenarios = 5 * 9 = 45
#

## Scenario settings
Scenario.name = %%Group.router%%_%%Group.bufferSize%%

Scenario.simulateConnections = false
Scenario.updateInterval = 5.0

# 43200s == 12h
Scenario.endTime = 43200
Scenario.nrofHostGroups = 1
Scenario.nrofHosts = 78


## Interface-specific settings:
# type : which interface class the interface belongs to
# For different types, the sub-parameters are interface-specific
# For SimpleBroadcastInterface, the parameters are:
# transmitSpeed : transmit speed of the interface (bytes per second)
# transmitRange : range of the interface (meters)

# "Bluetooth" interface for all nodes
btInterface.type = SimpleBroadcastInterface
# Transmit speed of 2 Mbps = 250kBps
btInterface.transmitSpeed = 250k
btInterface.transmitRange = 10


## Group-specific settings:

# Common settings for all groups
Group.movementModel = StationaryMovement
Group.nodeLocation = 10, 10

Group.bufferSize = [20M; 40M; 60M; 80M; 100M; 120M; 140M; 160M; 180M;]


# All nodes have the bluetooth interface
Group.nrofInterfaces = 1
Group.interface1 = btInterface
# Walking speeds
Group.speed = 0.5, 3
# Message TTL of 300 minutes (5 hours)
Group.msgTtl = 3000

Group.router = [EpidemicRouter; ProphetRouter; SprayAndWaitRouter; FirstContactRouter; DirectDeliveryRouter;]

SprayAndWaitRouter.nrofCopies = 16
SprayAndWaitRouter.binaryMode = true

ProphetRouter.secondsInTimeUnit = 30

Group.groupID = n
Group1.nrofHosts = 78



## Movement model settings
# seed for movement models' pseudo random number generator (default = 0)
MovementModel.rngSeed = 0
# World's size for Movement Models without implicit size (width, height; meters)
MovementModel.worldSize = 50, 50


# How many event generators
Events.nrof = 2

## Connection trace information
Events1.class = ExternalEventsQueue
Events1.filePath = my_scenarios/infocom-78n-24hr.tr

## Message creation events
Events2.class = MessageEventGenerator
Events2.hosts = 0,78
Events2.prefix = M
Events2.interval = 30,60
Events2.size = 50k,1M
Events2.time = 0,18000



## Reports - all report names have to be valid report classes
# how many reports to load
Report.nrofReports = 1
Report.reportDir = my_reports/buffer_size/
# Report classes to load
Report.report1 = MessageStatsReport


## Optimization settings -- these affect the speed of the simulation
## see World class for details.
Optimization.cellSizeMult = 5
Optimization.randomizeUpdateOrder = true

Now, let us take a look at some results. Figure 1 shows the delivery ratio of the messages obtained when these routing protocols were used together with the different buffer sizes.

Figure 1: Effects of buffer size on the delivery ratio.

 There are three clear trends in this Figure:

1. The delivery ratio steadily increased with the increasing buffer sizes in the case of Epidemic and PROPHET routing protocols.
2. For First Contact and Direct Delivery routing, the delivery ratio was invariant of the buffer sizes.
3. In case of SnW, the delivery ratio exhibited increasing trend upto a certain extent, beyond which it saturated.

Why?

The nature of the routing protocols provide some hints here. Unlimited replication-based benefit from the increasing buffer size since, more buffer space implies more nodes can carry with them more message copies. On the other hand, in case of single copy-based message forwarding, the same does not hold  true.

So, how does SnW behave with increasing buffer sizes for different values of L? You can verify it yourself. For different L, the increasing size of buffer helps improving the delivery ratio upto different extents beyond which the performance do not vary.

Check out the article Effects of Heterogeneity on the Performance of Pocket Switched Networks for a detailed discussion and many more interesting scenarios.