Parallel streaming signature EM-tree: A clustering algorithm for web scale applications

The proliferation of the web presents an unsolved problem of automatically analyzing billions of pages of natural language. We introduce a scalable algorithm that clusters hundreds of millions of web pages into hundreds of thousands of clusters. It does this on a single mid-range machine using efficient algorithms and compressed document representations. It is applied to two web-scale crawls covering tens of terabytes. ClueWeb09 and ClueWeb12 contain 500 and 733 million web pages and were clustered into 500,000 to 700,000 clusters. To the best of our knowledge, such fine grained clustering has not been previously demonstrated. Previous approaches clustered a sample that limits the maximum number of discoverable clusters. The proposed EM-tree algorithm uses the entire collection in clustering and produces several orders of magnitude more clusters than the existing algorithms. Fine grained clustering is necessary for meaningful clustering in massive collections where the number of distinct topics grows linearly with collection size. These fine-grained clusters show an improved cluster quality when assessed with two novel evaluations using ad hoc search relevance judgments and spam classifications for external validation. These evaluations solve the problem of assessing the quality of clusters where categorical labeling is unavailable and unfeasible.

Connected Viewing: Selling, Streaming and Sharing Media in the Digital Era

As patterns of media use become more integrated with mobile technologies and multiple screens, a new mode of viewer engagement has emerged in the form of connected viewing, which allows for an array of new relationships between audiences and media texts in the digital space. This exciting new collection brings together twelve original essays that critically engage with the socially-networked, multi-platform, and cloud-based world of today, examining the connected viewing phenomenon across television, film, video games, and social media. The result is a wide-ranging analysis of shifting business models, policy matters, technological infrastructure, new forms of user engagement, and other key trends affecting screen media in the digital era. Connected Viewing contextualizes the dramatic transformations taking place across both media industries and national contexts, and offers students and scholars alike a diverse set of methods and perspectives for studying this critical moment in media culture.

Infrastructure Disputes Shape Future of Streaming Media

In November 2010, tension between Internet infrastructure companies boiled over in a dispute between content distribution network (CDN) Level 3 and Internet service provider (ISP) Comcast. Level 3, a distribution partner of Netflix, accused Comcast of violating the principles of net neutrality when the ISP increased distribution fees for carrying high bandwidth services. Comcast justified its actions by stating that the price increase was standard practice and argued Level 3 was trying to avoid paying its fair share. The dispute exemplifies the growing concern over the rising costs of streaming media services. The companies facing these inflated infrastructure costs are CDNs (Level 3, Equinix, Limelight, Akamai, and Voxel), companies that host streaming media content on server farms and distribute the content to a variety of carriers, and ISPs (Comcast, Time Warner, Cox, and AT&T), the cable and phone companies that provide “last mile” service to paying customers. Both CDNs and ISPs are lobbying government regulators to keep their costs at a minimum. The outcome of these disputes will influence the cost, quality, and legal status of streaming media.

Real-World Extensions To Scheduling Algorithms Based On Lagrangian Relaxation

Recently, efficient scheduling algorithms based on Lagrangian relaxation have been proposed for scheduling parallel machine systems and job shops. In this article, we develop real-world extensions to these scheduling methods. In the first part of the paper, we consider the problem of scheduling single operation jobs on parallel identical machines and extend the methodology to handle multiple classes of jobs, taking into account setup times and setup costs, The proposed methodology uses Lagrangian relaxation and simulated annealing in a hybrid framework, In the second part of the paper, we consider a Lagrangian relaxation based method for scheduling job shops and extend it to obtain a scheduling methodology for a real-world flexible manufacturing system with centralized material handling.

Improving decisions for invasive species management: Reformulation and extensions of the Panetta-Lawes eradication graph

Aim: Effective decisions for managing invasive species depend on feedback about the progress of eradication efforts. Panetta & Lawes. developed the eradograph, an intuitive graphical tool that summarizes the temporal trajectories of delimitation and extirpation to support decision-making. We correct and extend the tool, which was affected by incompatibilities in the units used to measure these features that made the axes impossible to interpret biologically. Location: Victoria, New South Wales and Queensland, Australia. Methods: Panetta and Lawes' approach represented delimitation with estimates of the changes in the area known to be infested and extirpation with changes in the mean time since the last detection. We retain the original structure but propose different metrics that improve biological interpretability. We illustrate the methods with a hypothetical example and real examples of invasion and treatment of branched broomrape (Orobanche ramosa L.) and the guava rust complex (Puccinia psidii (Winter 1884)) in Australia. Results: These examples illustrate the potential of the tool to guide decisions about the effectiveness of search and control activities. Main conclusions: The eradograph is a graphical data summary tool that provides insight into the progress of eradication. Our correction and extension of the tool make it easier to interpret and provide managers with better decision support. © 2013 John Wiley & Sons Ltd.

Performance estimation of streaming applications for hierarchical MPSoCs

Parallel programming and effective partitioning of applications for embedded many-core architectures requires optimization algorithms. However, these algorithms have to quickly evaluate thousands of different partitions. We present a fast performance estimator embedded in a parallelizing compiler for streaming applications. The estimator combines a single execution-based simulation and an analytic approach. Experimental results demonstrate that the estimator has a mean error of 2.6% and computes its estimation 2848 times faster compared to a cycle accurate simulator.

Instability of Coupled Longitudinal and Transverse Electromagnetic Modes of Wave‐Propagation in Bounded Streaming Plasmas

The instability of coupled longitudinal and transverse electromagnetic modes associated with long wavelengths is studied in bounded streaming plasmas. The main conclusions are as follows: (i) For long waves for which O (k 2)=0, in the absence of relative streaming motion of electrons and ions and aωp/c<0.66, the whole spectrum of harmonic waves is excited due to finite temperature and boundary effects consisting of two subseries. One of these subseries can be identified with Tonks-Dattner resonance oscillations for the electrons, and arises primarily due to the electrons with frequencies greater than the electrostatic plasma frequency corresponding to the electron density in the midplane in the undisturbed state. The other series arises primarily due to ion motion. When aωp/c>0.66, in addition to the above spectrum of harmonic waves, the system admits an infinite number of growing and decaying waves. The instability associated with these modes is found to arise due to the interaction of the waves inside the plasma with the external electromagnetic field. (ii) For modes with comparatively shorter wavelengths for which O (k3)=0, the coupling due to finite temperature sets in, and it is found that the two series of harmonic waves obtained in (i) deriving energy from the transverse modes also become unstable. Thus, for these wavelengths the system admits three sets of growing and decaying modes, first two for all values of aωp/c and the third for (aωp/c) > 0.66. (iii) The presence of streaming velocities introduces various other coupling mechanisms, and we find that even for the wavelengths for which O (k2)=0, we get three sets of growing and decaying waves. The numerical values for the growth rates show that the streaming velocities enhance the growth rates of instability significantly.

Stochastic models of IEEE 802.11e wireless net works with multimedia applications

We provide a survey of some of our recent results ([9], [13], [4], [6], [7]) on the analytical performance modeling of IEEE 802.11 wireless local area networks (WLANs). We first present extensions of the decoupling approach of Bianchi ([1]) to the saturation analysis of IEEE 802.11e networks with multiple traffic classes. We have found that even when analysing WLANs with unsaturated nodes the following state dependent service model works well: when a certain set of nodes is nonempty, their channel attempt behaviour is obtained from the corresponding fixed point analysis of the saturated system. We will present our experiences in using this approximation to model multimedia traffic over an IEEE 802.11e network using the enhanced DCF channel access (EDCA) mechanism. We have found that we can model TCP controlled file transfers, VoIP packet telephony, and streaming video in the IEEE802.11e setting by this simple approximation.

Max-Coloring Paths: Tight Bounds and Extensions

The max-coloring problem is to compute a legal coloring of the vertices of a graph G = (V, E) with a non-negative weight function w on V such that Sigma(k)(i=1) max(v epsilon Ci) w(v(i)) is minimized, where C-1, ... , C-k are the various color classes. Max-coloring general graphs is as hard as the classical vertex coloring problem, a special case where vertices have unit weight. In fact, in some cases it can even be harder: for example, no polynomial time algorithm is known for max-coloring trees. In this paper we consider the problem of max-coloring paths and its generalization, max-coloring abroad class of trees and show it can be solved in time O(vertical bar V vertical bar+time for sorting the vertex weights). When vertex weights belong to R, we show a matching lower bound of Omega(vertical bar V vertical bar log vertical bar V vertical bar) in the algebraic computation tree model.

An SIMD machine for low-level vision

This paper presents an SIMD machine which has been tuned to execute low-level vision algorithms employing the relaxation labeling paradigm. Novel features of the design include: 1. (1) a communication scheme capable of window accessing under a single instruction. 2. (2) flexible I/O instructions to load overlapped data segments; and 3. (3) data-conditional instructions which can be nested to an arbitrary degree. A time analysis of the stereo correspondence problem, as implemented on a simulated version of the machine using the probabilistic relaxation technique, shows a speed up of almost N2 for an N × N array of PEs.

Parallel min-cut placement on reduced hardware SIMD architecture.

Massively parallel SIMD computing is applied to obtain an order of magnitude improvement in the executional speed of an important algorithm in VLSI design automation. The physical design of a VLSI circuit involves logic module placement as a subtask. The paper is concerned with accelerating the well known Min-cut placement technique for logic cell placement. The inherent parallelism of the Min-cut algorithm is identified, and it is shown that a parallel machine based on the efficient execution of the placement procedure.

Streaming FFT on REDEFINE-v2: An Application-Architecture Design Space Exploration

In this paper we explore an implementation of a high-throughput, streaming application on REDEFINE-v2, which is an enhancement of REDEFINE. REDEFINE is a polymorphic ASIC combining the flexibility of a programmable solution with the execution speed of an ASIC. In REDEFINE Compute Elements are arranged in an 8x8 grid connected via a Network on Chip (NoC) called RECONNECT, to realize the various macrofunctional blocks of an equivalent ASIC. For a 1024-FFT we carry out an application-architecture design space exploration by examining the various characterizations of Compute Elements in terms of the size of the instruction store. We further study the impact by using application specific, vectorized FUs. By setting up different partitions of the FFT algorithm for persistent execution on REDEFINE-v2, we derive the benefits of setting up pipelined execution for higher performance. The impact of the REDEFINE-v2 micro-architecture for any arbitrary N-point FFT (N > 4096) FFT is also analyzed. We report the various algorithm-architecture tradeoffs in terms of area and execution speed with that of an ASIC implementation. In addition we compare the performance gain with respect to a GPP.

On the Implementation of a Streaming Video over Peer to Peer network using Middleware Components

Video streaming applications have hitherto been supported by single server systems. A major drawback of such a solution is that it increases the server load. The server restricts the number of clients that can be simultaneously supported due to limitation in bandwidth. The constraints of a single server system can be overcome in video streaming if we exploit the endless resources available in a distributed and networked system. We explore a P2P system for streaming video applications. In this paper we build a P2P streaming video (SVP2P) service in which multiple peers co-operate to serve video segments for new requests, thereby reducing server load and bandwidth used. Our simulation shows the playback latency using SVP2P is roughly 1/4th of the latency incurred when the server directly streams the video. Bandwidth consumed for control messages (overhead) is as low as 1.5% of the total data transfered. The most important observation is that the capacity of the SVP2P grows dynamically.

Effects of acoustic-streaming-induced flow in evaporating nanofluid droplets

We study the effect of acoustic streaming on nanoparticle motion and morphological evolution inside an acoustically levitated droplet using an analytical approach coupled with experiments. Nanoparticle migration due to internal recirculation forms a density stratification, the location of which depends on initial particle concentration. The time scale of density stratification is similar to that of perikinetic-driven agglomeration of particle flocculation. The density stratification ultimately leads to force imbalance leading to a unique bowl-shaped structure. Our analysis shows the mechanism of bowl formation and how it is affected by particle size, concentration, internal recirculation and fluid viscosity.