Recently we could observe a huge change in the mobile industry when the original idea of mobile phone was
transformed into the new concept of mobile multimedia devices capable to perform multiple complex tasks and
integrating a number of functionalities. As a consequence it resulted in significant increase of the device integration time
and cost and complicated deployment of the new technologies. The device integrators are forced to favor modularity
everywhere where it is possible in design of new devices, which results in a new trend towards networked architectures
for the mobile devices.
However, moving towards networked architectures specifically designed to overcome limitations brought by the mobile
devices is a time consuming task. It requires fresh mind analysis of many solutions applied in other contexts, since some
of the constraints and requirements are unique in comparison with e.g. SoC, NoC, which are the most known embedded
network solutions, and of course they are significantly different comparing to the wide area networks. The main
differentiating factors are: strongly constrained power consumption by the battery life time; and a need for modular
architecture to allow reuse of the existing components or modules.
The paper provides an overview of the state of art in the embedded networks research and describes general background
for our studies, key assumptions, restrictions and limitations that we faced at the beginning of development of the
embedded networks architecture for mobile devices.
The paper describes new method for routing classification of the packets in the networks with static routing. Static
routing is mostly used in the stable and well-controlled networks, which are sensitive to the additional complexity
introduced by the dynamic routing schemes. For example, it is a straightforward choice for the embedded networks on
terminal. The available solutions for the static routing require full-size or longest prefix-based aggregated routing table,
which is inefficient taking into account static nature of the performed routing decisions. The standard approach is based
on the full implementation of the route lookup procedure, which for every packet performs search for the longest prefix
match in the routing table. As a result, it increases implementation complexity of the network end points and switches,
requires high speed memory for storing routing tables, and leads to an additional processing delay and energy
consumption for each transmitted packet.
This work is based on the observation that clever planning of the networks with static routing allows applying new
principles of route lookup. As a consequence it allows significantly reduce complexity of the packet classification and
forwarding procedures and minimize amount of consumed resources. Complexity reduction of the packets classification
and forwarding procedures allows simplifying implementation of the networking part of the protocol stack, which results
in reduction of the device components cost, decrease of the power and memory consumption, as well as the packet
The paper contains a description of the proposed route classification method and discusses its applicability for
broad range of networks with static routing.
The network stability, performance and QoS guaranties given to the end users strongly depend on the routing protocol. A significant increase of the routing tables average size degrades the routing protocol performance. The route aggregation can be used for decreasing the amount of processing and storing routing information, but creates a lot of configuration and control problems. Traditionally, the route aggregation is used for a fixed network configuration. The edge routers process and store the lower level network detail but only pass the aggregate information to the upper level. In this way, the edge routers saving the upper level routers from unnecessary details and also deny other edge routers the detailed information needed to calculate an efficient aggregation plan for all lower level networks. Later exceptions to the address plan can be handled by CIDR principles, but at the cost of inserting the exception detail into the upper level network to be processed and stored by all of its routers.
The paper describes how routers can supplement routing information to automatically determine the best route aggregation and does not require initial configuration of aggregation rules, but instead, only the network interfaces to the subnetworks. Aggregation is thus no longer statically constrained by and pre-calculated from the initial network addressing plan, but automatic and flexible to changes as the routing protocol itself is. The supplemental information needed to do this may be exchanged by using extensions of the routing protocol or a separate message exchange protocol installed on the edge routers of the network.
Traditionally dynamic load balancing is applied in resource-reserved connection-oriented networks with a large degree of managed control. Load balancing in connectionless networks is rather rudimentary and is either static or requires network-wide load information. This paper presents a fully automated, traffic driven dynamic load balancing mechanism that uses local load information. The proposed mechanism is easily deployed in a multi-vendor environment in which only a subset of routers supports the function.
The Dynamic Localized Load Balancing (DLLB) mechanism distributes traffic based on two sets of weights. The first set is fixed and is inverse proportional to the path cost, typically the sum of reciprocal bandwidths along the path. The second weight reflects the utilization of the link to the first next hop along the path, and is therefore variable. The ratio of static weights defines the ideal load distribution, the ratio of variable weights the node-local load distribution estimate. By minimizing the difference between variable and fixed ratios the traffic distribution, with the available node-local knowledge, is optimal. The above mechanism significantly increases throughput and decreases delay from a network-wide perspective. Optionally the variable weight can include load information of nodes downstream to prevent congestion on those nodes. The latter function further improves network performance, and is easily implemented on top of the standard OSPF signaling. The mechanism does not require many node resources and can be implemented on existing router platforms.
We have developed a closed equation that allows calculation of the optimal bandwidth based on estimates of volume and quality of service criteria of user traffic. The basic assumption underlying the equation is that traffic to be dropped at a bottleneck link should not be sent to that link in the first place. Instead, such traffic should be dropped as early as possible. The equation was validated with simulations using a topology and a traffic mix typical for 3G radio access networks. The quality of service of user traffic to the routers on either side of the backup link was considerably better compared to standard methods. The equation uses only a few parameters based on the traffic demand matrix and is easily incorporated in a network-planning tool to calculate backup links or bandwidth requirements for logical pipes. It can also be used as basis for complete network planning.
During the transient period after a link failure the network cannot guarantee the agreed service levels to user data. This is due to the fact that forwarding tables in the network are inconsistent. Moreover, link states can inadvertently be advertised wrong due to protocol time outs, which may result in persistent route flaps. Reducing the probability of wrongly advertised link states, and the time during which the forwarding tables are inconsistent, is therefore of eminent importance to provide consistent and high level QoS to user data.
By queuing routing traffic in a queue with strict priority over all other (data) queues, i.e. assigning the highest priority in a Differentiated Services model, we were able to reduce the probability of routing data loss to almost zero, and reduce flooding times almost to their theoretical limit. The quality of service provided to user traffic was considerable higher than without the proposed modification.
The scheme is independent of the routing protocol, and can be used with most differentiated service models. It is compatible with the current OSPF standard, and can be used in conjunction with other improvements in the protocol with similar objectives.