Fatemeh Rahimian

JA-BE-JA: A Distributed Algorithm for Balanced Graph Partitioning

Balanced graph partitioning is a well known NP-complete problem with a wide range of applications. These applications include many large-scale distributed problems including the optimal storage of large sets of graph-structured data over several hosts-a key problem in todays Cloud infrastructure. However, in very large-scale distributed scenarios, state-of-the-art algorithms are not directly applicable, because they typically involve frequent global operations over the entire graph. In this paper, we propose a fully distributed algorithm, called JA-BE-JA, that uses local search and simulated annealing techniques for graph partitioning. The algorithm is massively parallel: there is no central coordination, each node is processed independently, and only the direct neighbors of the node, and a small subset of random nodes in the graph need to be known locally. Strict synchronization is not required. These features allow JA-BE-JA to be easily adapted to any distributed graph-processing system from data centers to fully distributed networks. We perform a thorough experimental analysis, which shows that the minimal edge-cut value achieved by JA-BE-JA is comparable to state-of-the-art centralized algorithms such as METIS. In particular, on large social networks JA-BEJA outperforms METIS, which makes JA-BE-JA-a bottom-up, self-organizing algorithm-a highly competitive practical solution for graph partitioning.

Vitis: A Gossip-based Hybrid Overlay for Internet-scale Publish/Subscribe Enabling Rendezvous Routing in Unstructured Overlay Networks

Peer-to-peer overlay networks are attractive solutions for building Internet-scale publish/subscribe systems. However, scalability comes with a cost: a message published on a certain topic often needs to traverse a large number of uninterested (unsubscribed) nodes before reaching all its subscribers. This might sharply increase resource consumption for such relay nodes (in terms of bandwidth transmission cost, CPU, etc) and could ultimately lead to rapid deterioration of the systems performance once the relay nodes start dropping the messages or choose to permanently abandon the system. In this paper, we introduce {\em Vitis}, a gossip-based publish/subscribe system that significantly decreases the number of relay messages, and scales to an unbounded number of nodes and topics. This is achieved by the novel approach of enabling rendezvous routing on unstructured overlays. We construct a hybrid system by injecting structure into an otherwise unstructured network. The resulting structure resembles a navigable small-world network, which spans along clusters of nodes that have similar subscriptions. The properties of such an overlay make it an ideal platform for efficient data dissemination in large-scale systems. We perform extensive simulations and evaluate Vitis by comparing its performance against two base-line publish/subscribe systems: one that is oblivious to node subscriptions, and another that exploits the subscription similarities. Our measurements show that Vitis significantly outperforms the base-line solutions on various subscription and churn scenarios, from both synthetic models and real-world traces.

gradienTv: market-based P2P live media streaming on the gradient overlay

This paper presents gradienTv, a distributed, market-based approach to live streaming. In gradienTv, multiple streaming trees are constructed using a market-based approach, such that nodes with increasing upload bandwidth are located closer to the media source at the roots of the trees. Market-based approaches, however, exhibit slow convergence properties on random overlay networks, so to facilitate the timely discovery of neighbours with similar upload bandwidth capacities (thus, enabling faster convergence of streaming trees), we use the gossip-generated Gradient overlay network. In the Gradient overlay, nodes are ordered by a gradient of node upload capacities and the media source is the highest point in the gradient. We compare gradienTv with state-of-the-art NewCoolstreaming in simulation, and the results show significantly improved bandwidth utilization, playback latency, playback continuity, and reduction in the average number of hops from the media source to nodes.

Sepidar: Incentivized Market-Based P2P Live-Streaming on the Gradient Overlay Network

Live streaming of video content using overlay networks has gained widespread adoption on the Internet. This paper presents Sepidar, a distributed market-based model, that builds and maintains overlay network trees, which are approximately minimal height, for delivering live media as a number of sub streams. A streaming tree is constructed for each sub stream such that nodes that contribute higher amounts of upload bandwidth are located increasingly closer to the media source at the root of the tree. While our distributed market model can be run against a random sample of nodes, we improve its convergence time to stabilize a tree by executing against a sample of nodes that contribute similar amounts of upload bandwidth. We use the Gradient overlay network to generate samples of such nodes. We address the problem of free-riding through parent nodes auditing the behaviour of their child nodes. We evaluate Sepidar by comparing it in simulation with state-of-the-art New Cool streaming. Our results show significantly improved playback latency and playback continuity under churn, flash crowd, and catastrophic failure experiment scenarios. We also show that using the Gradient improves convergence time of our distributed market model compared to a random overlay network. Finally, we show that Sepidar punishes the performance of free-riders, and that nodes are incentivized to contribute more upload bandwidth by relatively improved performance.

Distributed Vertex-Cut Partitioning

Graph processing has become an integral part of big data analytics. With the ever increasing size of the graphs, one needs to partition them into smaller clusters, which can be managed and processed more easily on multiple machines in a distributed fashion. While there exist numerous solutions for edge-cut partitioning of graphs, very little effort has been made for vertex-cut partitioning. This is in spite of the fact that vertex-cuts are proved significantly more effective than edge-cuts for processing most real world graphs. In this paper we present Ja-be-Ja-vc, a parallel and distributed algorithm for vertex-cut partitioning of large graphs. In a nutshell, Ja-be-Ja-vc is a local search algorithm that iteratively improves upon an initial random assignment of edges to partitions. We propose several heuristics for this optimization and study their impact on the final partitioning. Moreover, we employ simulated annealing technique to escape local optima. We evaluate our solution on various graphs and with variety of settings, and compare it against two state-of-the-art solutions. We show that Ja-be-Ja-vc outperforms the existing solutions in that it not only creates partitions of any requested size, but also requires a vertex-cut that is better than its counterparts and more than 70% better than random partitioning.

Locality-Awareness in a peer-to-peer publish/subscribe network

Peer-to-peer publish/subscribe systems are promising solutions to provide distributed content distribution services at Internet-scale with low cost. One of the potential problems with peer-to-peer overlays, however, is the inefficient traffic and large delays, due to the mismatch between the physical network and the overlay topology. This paper introduces a locality-aware extension to a peer-to-peer publish/subscribe system, named Vitis. The ultimate purpose is to avoid communications over long-distance links, instead, nodes send data over short-distance and low-cost links, when possible, while maintaining an acceptable quality of service. We show, through simulations, that the average data delivery time is up to 40% improved. The cost to pay is at most 10% more relaying in the peer-to-peer overlay.

A Distributed Algorithm for Large-Scale Graph Partitioning

Balanced graph partitioning is an NP-complete problem with a wide range of applications. These applications include many large-scale distributed problems, including the optimal storage of large sets of graph-structured data over several hosts. However, in very large-scale distributed scenarios, state-of-the-art algorithms are not directly applicable because they typically involve frequent global operations over the entire graph. In this article, we propose a fully distributed algorithm called JA-BE-JA that uses local search and simulated annealing techniques for two types of graph partitioning: edge-cut partitioning and vertex-cut partitioning. The algorithm is massively parallel: There is no central coordination, each vertex is processed independently, and only the direct neighbors of a vertex and a small subset of random vertices in the graph need to be known locally. Strict synchronization is not required. These features allow JA-BE-JA to be easily adapted to any distributed graph-processing system from data centers to fully distributed networks. We show that the minimal edge-cut value empirically achieved by JA-BE-JA is comparable to state-of-the-art centralized algorithms such as Metis. In particular, on large social networks, JA-BE-JA outperforms Metis. We also show that JA-BE-JA computes very low vertex-cuts, which are proved significantly more effective than edge-cuts for processing most real-world graphs.

Boosting Vertex-Cut Partitioning for Streaming Graphs

While the algorithms for streaming graph partitioning are proved promising, they fall short of creating timely partitions when applied on large graphs. For example, it takes 415 seconds for a state-of-the-art partitioner to work on a social network graph with 117 millions edges. We introduce an efficient platform for boosting streaming graph partitioning algorithms. Our solution, called HoVerCut, is Horizontally and Vertically scalable. That is, it can run as a multi-threaded process on a single machine, or as a distributed partitioner across multiple machines. Our evaluations, on both real-world and synthetic graphs, show that HoVerCut speeds up the process significantly without degrading the quality of partitioning. For example, HoVerCut partitions the aforementioned social network graph with 117 millions edges in 11 seconds that is about 37 times faster.

Smart partitioning of geo-distributed resources to improve cloud network performance

Cloud Computing systems with geo-distributed resources are becoming more popular for enabling a new family of applications, which are latency sensitive or bandwidth intensive, e.g., Internet of Things and online video gaming services. The approach is to host the cloud services at the network edges to reduce the latency and bandwidth consumption. However, the topology of the existing networks is not necessarily optimal for hosting Cloud services. Moreover, how the services are placed on the nodes, can affect the performance of the applications and the whole network. Therefore, we propose a novel algorithm to partition a distributed infrastructure into a set of computing clusters, each called a Micro Data Center. Our proposed algorithm is a decentralized community detection algorithm that does not require any global knowledge of the network topology. We compare our solution with a geolocation based clustering solution and demonstrate our preliminary results based on a real world network data set. We show that micro data centers increase the minimum available bandwidth in the network to up to 62%. Likewise, the average latency can be reduced to 50%.

DeGPar: Large Scale Topic Detection Using Node-Cut Partitioning on Dense Weighted Graphs

Topic Detection (TD) refers to automatic techniques for locating topically related material in web documents. Nowadays, massive amounts of documents are generated by users of Online Social Networks (OSNs), in form of very short text, tweets and snippets of news. While topic detection, in its traditional form, is applied to a few documents containing a lot of information, the problem has now changed to dealing with massive number of documents with very little information. The traditional solutions, thus, fall short either in scalability (due to huge number of input items) or sparsity (due to insufficient information per input item). In this paper we address the scalability problem by introducing an efficient and scalable graph based algorithm for TD on short texts, leveraging dimensionality reduction and clustering techniques. We first, compress the input set of documents into a dense graph, such that frequent cooccurrence patterns in the documents create multiple dense topological areas in the graph. Then, we partition the graph into multiple dense sub-graphs, each representing a topic. We compare the accuracy and scalability of our solution with two state-of-the-art solutions (including the standard LDA, and BiTerm). The results on two widely used benchmark datasets show that our algorithm not only maintains a similar or better accuracy, but also performs by an order of magnitude faster than the state-of-the-art approaches.