Mohammad Asad R. Chaudhry

On the Minimum Number of Transmissions in Single-Hop Wireless Coding Networks

The advent of network coding presents promising opportunities in many areas of communication and networking. It has been recently shown that network coding technique can significantly increase the overall throughput of wireless networks by taking advantage of their broadcast nature. In wireless networks, each transmitted packet is broadcasted within a certain area and can be overheard by the neighboring nodes. When a node needs to transmit packets, it employs the opportunistic coding approach that uses the knowledge of what the nodes neighbors have heard in order to reduce the number of transmissions. With this approach, each transmitted packet is a linear combination of the original packets over a certain finite field. In this paper, we focus on the fundamental problem of finding the optimal encoding for the broadcasted packets that minimizes the overall number of transmissions. We show that this problem is NP-complete over GF(2) and establish several fundamental properties of the optimal solution. We also propose a simple heuristic solution for the problem based on graph coloring and present some empirical results for random settings.

Efficient algorithms for Index Coding

The index coding problem has attracted a considerable amount of attention in recent years. The problem is motivated by several applications in wireless networking and distributed computing, including wireless architectures that utilize network coding and opportunistic listening. In this paper, we propose efficient exact and heuristic solutions for this problem. Our numerical study shows that exact solutions can be efficiently obtained for small instances of the problem, while heuristic solutions with low computation time can achieve near-optimal performance for large instances.

On the complementary Index Coding problem

The Index Coding problem is one of the basic problems in wireless network coding. In this problem, a server needs to deliver a set P of packets to several clients through a noiseless broadcast channel. Each client needs to obtain a certain subset of P and has prior side information about a different subset of P. The objective is to satisfy the requirements of all clients with the minimum number of transmissions. Recently, it was shown that the Index Coding problem is NP-hard. Furthermore, this problem was shown to be hard to approximate under a widely accepted complexity assumption.

A Two-Way Street: Green Big Data Processing for a Greener Smart Grid

Integrating renewables into the mainstream energy market is pivotal for the green revolution promised by the smart grid. The real power behind realization of the smart grid goals lies in the volume, variety, velocity of the big data generated by a variety of sources. Nevertheless, the smart grid needs data centers to digest the big data for its profound green revolution. However, big data processing is the radix for data centers to be seen as energy black holes. Unless data centers are transformed into energy-efficient enterprises, big data are going to be responsible for superfluous energy burn, potentially reversing the smart grid genesis with regard to green environmental impact. This paper describes the role of the big data enterprise in envisioning the smart grid. We dissect the big data enterprise into six vital planes impacting the energy footprints of data centers. We present a survey of key strategies to make these six vital planes greener. Moreover, we present open challenges and directions in this context. We assert that a cross-plane approach toward a greener optimization is crucial. In this vein, we present a green orchestrator that is capable of incorporating different planes in an integrated fashion to boost energy profile of the big data enterprise.

Finding Sparse Solutions for the Index Coding Problem

The Index Coding problem has recently attracted a significant attention from the research community. In this problem, a server needs to deliver data to a set of wireless clients over the broadcast channel. Each client requires one or more packets, but it might have access to the packets requested by other clients as side information. The goal is to deliver the required data to each client with minimum number of transmissions. In this paper, we focus on finding sparse solutions to the Index Coding problem. In a sparse solution each transmitted packet is a linear combination of at most two original packets. We focus both on scalar and vector versions of the problem. For the scalar case, we present a polynomial time algorithm that achieves an approximation ratio of 2-(1/¿an). For the vector case, we present a polynomial time algorithm that identifies an optimal solution to the problem. Our simulation studies demonstrate that our algorithms achieve good performance in practical scenarios.

Greener Data Exchange in the Cloud: A Coding-Based Optimization for Big Data Processing

The rise of the cloud and distributed data-intensive (big data) applications puts pressure on data center networks due to the movement of massive volumes of data. Reducing the volume of communication is pivotal for embracing greener data exchange by efficient utilization of network resources. This paper proposes the use of mixing technique, spate coding, working in tandem with software-defined network control as a means of dynamically-controlled reduction in volume of communication. We introduce motivating real-world use-cases, and present a novel spate coding algorithm for the data center networks. We also analyze the computational complexity of the general problem of minimizing the volume of communication in a distributed data center application without degrading the rate of information exchange, and provide theoretical limits of such schemes. Moreover, we proceed to bridge the gap between theory and practice by performing a proof-of-concept implementation of the proposed system in a real world data center. We use Hadoop MapReduce, the most widely used big data processing framework, as our target. The experimental results employing two of industry standard benchmarks show the advantage of our proposed system compared to a vanilla Hadoop implementation, an in-network combiner, and Combine-N-Code. The proposed coding-based scheme shows performance improvement in terms of volume of communication (up to 62%), goodput (up to 76%), disk utilization (up to 38%), and the number of bits that can be transmitted per Joule of energy (up to 200%).

CodHoop: A system for optimizing big data processing

The rise of the cloud and distributed data-intensive (“Big Data”) applications puts pressure on data center networks due to the movement of massive volumes of data. This paper proposes CodHoop a system employing network coding techniques, specifically index coding, as a means of dynamically-controlled reduction in volume of communication. Using Hadoop as a representative of this class of applications, a motivating use-case is presented. The proof-of-concept implementation results exhibit an average advantage of 31% compared to vanilla Hadoop implementation which depending on use-case translates to 31% less energy utilization of the equipment, 31% more jobs that run simultaneously, or to a 31% decrease in job completion time.

Efficient Rerouting Algorithms for Congestion Mitigation

Congestion mitigation and overflow avoidance are two of the major goals of the global routing stage. With a significant increase in the chip size and routing complexity,congestion and overflow have become critical issues in physical design automation. In this paper we present several routing algorithms for congestion reduction and overflow avoidance.Our methods are based on ripping up nets that go through the congested regions and replacing them with congestion-aware Steiner trees. We propose several efficient algorithms for finding congestion-aware Steiner trees and evaluate their performance using the ISPD routing benchmarks. We also show that the novel technique of network coding contributes to further improvements in routability and reduction of congestion. Accordingly, we propose an algorithm for identifying efficient congestion-aware network coding topologies and evaluate its performance.

Efficient congestion mitigation using congestion-aware steiner trees and network coding topologies

In the advent of smaller devices, a significant increase in the density of on-chip components has raised congestion and overflow as critical issues in VLSI physical design automation. In this paper, we present novel techniques for reducing congestion and minimizing overflows. Our methods are based on ripping up nets that go through the congested areas and replacing them with congestion-aware topologies. Our contributions can be summarized as follows. First, we present several efficient algorithms for finding congestion-aware Steiner trees that is, trees that avoid congested areas of the chip. Next, we show that the novel technique of network coding can lead to further improvements in routability, reduction of congestion, and overflow avoidance. Finally, we present an algorithm for identifying efficient congestion-aware network coding topologies. We evaluate the performance of the proposed algorithms through extensive simulations.

Efficient Network Coding Algorithms for Dynamic Networks

Network coding is a new paradigm that allows the intermediate nodes in a network to create new packets by combining the packets received on their incoming edges. The central problem in the design of network coding schemes is to assign local encoding coefficients for the intermediate edges in a way that allows every terminal node to decode the packets generated by the source node. The main applications of the network coding technique include content distribution, peer-to-peer networks, and wireless ad-hoc networks. Such networks are characterized by highly dynamic set of users and frequent topological changes. In this paper we focus on the design of efficient multicast network codes for dynamic networks. First, we consider the problem of maintaining the feasibility of a given network code upon a change in the network topology or the addition of a new user. Our goal is to minimize the number of encoding coefficients that needs to be modified to keep the network code feasible. Second, we present a new network coding algorithm that uses path-based coding assignments to efficiently handle frequent changes in the network topology and the multicast group.