Gautam Barua

Tracing attackers with deterministic edge router marking (DERM)

Tracing the attackers in a distributed denial-of-service(DDoS) attack is particularly difficult since attackers spoof the source addresses We present a novel approach to IP Traceback – Deterministic Edge Router Marking (DERM) The proposed scheme is scalable to thousands of attackers, is very simple to implement at the routers, has no bandwidth overhead and needs minimal processing and storage requirements at the victim As each complete mark fits into a single packet, our scheme can also be used for per-packet filtering and as a congestion signature in a pushback protocol The traceback procedure requires a small number of packets and can be performed during the post-mortem analysis of an attack Only limited co-operation is required from Internet Service Providers (ISP) They do not have to reveal the topology of their internal networks.

Performance comparison of two on-demand routing protocols for mobile ad hoc networks

AODV and DSR are two of the most important routing protocols for ad hoc networks. In this article, we make a comparison of these routing protocols in context of the services offered by them to the transport layer. We first study the interactions between the Transport and MAC layer and then, based on observations, propose metrics to measure the effectiveness of the two routing protocols. Detailed simulations are then performed to compare the performance of AODV and DSR under varying load, mobility and network size and connectivity. Our work makes clear the need for routing protocols to consider network load as an important factor in addition to path length while computing routes. In addition, we find that a suitable caching mechanism is necessary for scalability of a routing protocol and that its effectiveness increases with increasing connection density and decreasing mobility.

Adaptive routing for ad hoc wireless networks providing QoS guarantees

A cluster-based route discovery and dynamic route management protocol for ad hoc networks with inaccurate information for given QoS (quality of service) requirements has been proposed and compared with existing routing protocols. The proposed scheme is much more scalable, the route discovery is faster (O(logN) against O(N), N being the number of nodes) and it also guarantees QoS - a feature which has not been incorporated in any existing protocol. Also, features such as local dynamic route maintenance, loop avoidance load balancing, delay bounded routing and guaranteed rate routing have been included in the architecture of ACRQ. ACRQ can also work in case of inaccurate information by assigning weights according to the probability of each information being true. This information is then used for routing decisions. Furthermore, ACRQ takes advantage of the MAC characteristics of present wireless technologies such as master-driven communication in Bluetooth. Master-driven systems encourage cluster based routing, which is also helpful for scalability. A distributed algorithm for clustering and election of cluster heads (in ACRQ protocol) with only one round of message exchange has been proposed as well.

On-line consistent backup in transactional file systems

A consistent backup, preserving data integrity across files in a file system, is of utmost importance for the purpose of correctness and minimizing system downtime during the process of data recovery. With the present day demand for continuous access to data, backup has to be taken of an active file system, putting the consistency of the backup copy at risk. We propose a scheme to take a consistent backup of an active file system assuming that the file system supports transactions. We take into consideration that file operations include, besides reading and writing of files, directory operations, file descriptor operations and operations such as append, truncate, etc. We are putting our scheme into a formal framework to prove its correctness and we have devised algorithms for implementing the scheme. We intend to simulate the algorithms and run experiments to help tune the algorithms.

Enhanced algorithms for consistent network updates

The basic algorithm that consistently updates the switches of a Software Defined Network while preserving the update property of per-packet consistency is the two-phase update. The two-phase update is underspecified on two matters: 1) how to detect when the last packet of the old rule set has left the network and therefore exactly when to delete the old rules 2) recoverability of the update. Recoverability ensures that those updates for which the two-phase algorithm is not completely executed do not change the semantics of those updates for which the algorithm is completely executed. This paper examines the failures that can occur during an update and how to handle a subset of those failures. It proposes an algorithm, enhancing the two-phase update to handle deletion of old rules and recoverability. It also specifies how to effectively use a software cache to supplement the TCAM, during an update. The paper extends the same algorithm for per-flow consistent updates, using a restricted number of exact-match rules for existing flows and specifying when to delete the exact-match rules. The two algorithms are also specified for switches that do not support a software cache. The paper also analyzes the algorithms quantitatively, identifying the parameters of interest and what they depend upon.

Load balancing inbound traffic in multihomed stub autonomous systems

In the Internet, with many competing networks each trying to optimise its own bandwidth, a stub network has limited knowledge about user demands, available network resources and routing policies of other networks. This uncertainty makes the task of interdomain traffic engineering for a stub network very challenging. The basic aim of a stub network connected to multiple ISPs (multihomed) is to load balance its traffic among its various edge links. Our goal in this work is to distribute the incoming traffic of a multihomed stub network among its various edge links. The focus is on networks that primarily download traffic from the Internet. Regulating the incoming traffic is difficult since it will require to influence the behaviour of the remote destinations. We performed a systematic analysis of our problem and showed that even a restricted instance of the problem is NP-complete. We proposed simple, low-cost route control techniques to load balance traffic by reallocating the routes of outgoing traffic. The techniques were validated using synthetic as well as actual data collected under numerous traffic load conditions. Results show that we can achieve significant improvement in load balancing with minimum traffic re-assignments. Moreover, the proposed techniques neither require any third party assistance nor changes to existing protocols and network setup. This makes our schemes easily deployable in real networks.

Pricing with Bandwidth Guarantees for Clients with multi-ISP Connections

The increase in internet coverage and decrease in internet access price has resulted in demand for good internet service. Clients want some guarantee in internet access quality. In this paper, we present a model in which clients are guaranteed connection and bandwidth and if clients do not get the service they request, the service provider pays a penalty to the clients. We consider a system of internet clients with multiple internet service provider (ISP) connections to a set of ISPs. When a client arrives, an ISP has to decide whether to accept the client, and then the price to charge from the client for the duration of its connection. Rejection of a client results in a penalty and delay in getting the requested bandwidth also incurs a penalty. We assume a Poisson arrival process with the rate of arrival sensitive to the price being charged. A client requests bandwidth for a time that is exponentially distributed, then the client is idle for a time that is also exponentially distributed; and then either the client departs or requests bandwidth again after the idle period is over. A service provider tries to maximize its income by charging appropriate prices based on its current state and deciding whether to accept more clients or not. Since penalties are imposed, such solutions also automatically balance load among service providers, and so the quality of service to clients improves. We present solutions that maximize the income of service providers. The solutions are then compared using simulation. Simulation results show that our analysed solution significantly improve quality of service of clients and increase the income of service providers as compared to a simple heuristic based solution that is otherwise could to be used.

Consistent Online Backup in Transactional File Systems

The backup taken of a file system must be consistent, preserving data integrity across files in the file system. With file system sizes getting very large, and with demand for continuous access to data, backup has to be taken when the file system is active (is online). Arbitrarily taken online backup may result in an inconsistent backup copy. We propose a scheme referred to as mutual serializability to take a consistent backup of an active file system assuming that the file system supports transactions. The scheme extends the set of conflicting operations to include read-read conflicts, and it is shown that if the backup transaction is mutually serializable with every other transaction individually, a consistent backup copy is obtained. The user transactions continue to serialize within themselves using some standard concurrency control protocol such as Strict 2PL. We put our scheme into a formal framework to prove its correctness, and the formalization as well as the correctness proof are independent of the concurrency control protocol used to serialize user transactions. The scheme has been implemented and experiments show that consistent online backup is possible with reasonable overhead.

Distributed route control schemes to load balance incoming traffic in multihomed stub networks

Load balancing traffic of a multihomed stub network among its various egress links can significantly improve performance by allowing the network to route its traffic around congested paths. In [3], several centralized route control schemes to load balance traffic were proposed. It was shown that employing these route control schemes resulted in significant improvement in round-trip times. We carry this exploration further by developing distributed algorithms to load balance traffic. We show theoretically that the expected performance of the distributed version is the same as the centralized routing scheme. We further validate the schemes using both synthetic as well as actual traffic traces.

Effective Bandwidth Utilisation in Multihoming Networks

In order to bridge the speed disparity at the LAN-WAN interface, enterprises install a number of policy-based devices to enforce administrative policies over the WAN access. The conventional techniques of bandwidth management concentrate only on the outgoing traffic from the local network, since that is the only traffic that can be controlled. However it is usually the incoming traffic that causes bottlenecks. This paper proposes a simple load balancing approach to optimize the use of WAN links in a multihomed environment. Incoming traffic on a link is controlled by diverting outgoing traffic from some nodes to other links. Incoming bandwidth is measured, thresholds are defined, and traffic diversion takes place to maintain the thresholds. We present a scheme to implement this. A prototype implementation has been done on an educational campus network. Preliminary experiments have shown that there are significant improvements in performance. These results are presented. Proposed future work to extend the scheme to an ISP network is described