1 / 10

Interactive Parallel Data Visualization and Exploration

Interactive Parallel Data Visualization and Exploration. March 27, 2013 Mihai Budiu Moises Goldszmidt Jean-Philippe Martin Mark Manasse Alex Andoni Gordon Plotkin Qingzhuo Luo (intern 2012 ). Cluster-Based Visualization. Data. RMI. Partial renderings. Rendering overlay.

fraley
Download Presentation

Interactive Parallel Data Visualization and Exploration

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Interactive Parallel Data Visualization and Exploration March 27, 2013 Mihai Budiu Moises Goldszmidt Jean-Philippe Martin Mark Manasse Alex Andoni Gordon Plotkin Qingzhuo Luo (intern 2012)

  2. Cluster-Based Visualization Data RMI Partial renderings Renderingoverlay Client GUI Multi-core data extraction, transformationrendering Server workers

  3. Data size O(1) RMI Partial renderings O(n) O(1) Renderingoverlay ~1MPx = O(1) Client GUI Multi-core data extraction, transformationrendering Server workers

  4. Composable Interfaces Client IView Cluster-level parallelism ParallelView IView IView IView ProxyView ProxyView LocalView IView IView Multi-core-level parallelism ParallelView ParallelView IView IView IView IView IView LocalView LocalView LocalView LocalView LocalView Server 1 Server 2

  5. Computing “Small” Results Plot Image overlay IDistribution ParallelDistr Plot Image Result size independent of the data size IDistribution ProxyDistr Client Image overlay Plot Server IDistribution ParallelDistr Image Image Plot Plot IDistribution IDistribution LocalDistr LocalDistr

  6. Computing “Big” Results Filter result IDistribution ParallelDistr IDistribution ParallelDistr Filter IDistribution ProxyDistr IDistribution Data nevermoves ProxyDistr Client Filter Server IDistribution ParallelDistr IDistribution ParallelDistr Filter Filter IDistribution IDistribution LocalDistr LocalDistr IDistribution IDistribution LocalDistr LocalDistr

  7. A Language of Linear Transformations • f : A → B, f(a + b) = f(a) + f(b) • Various + operations: • DB Tables, with union • Distributions, with union • Histograms, with pointwise addition • Transparent images with overlays • Timeseries, with union • Multisets, with pointwise addition • Top K values, with mergesort

  8. Software Stack server side client side Domain-specific App Domain-specific Data-structures Plotting Built-in structs (Distribution, View, Histogram, ScatterPlot, TimeSeries, ColorMap) IView IDistribution IHistogram IScatter ITimeSeries IColorMap IPlot X ILocal IParallel IProxy RMI Object transport

  9. Running on 700 cores

More Related