Live Reorderable Accordion Drawing LiveRAC

1
Live Reorderable Accordion Drawing (LiveRAC)

• Peter McLachlan
• MSc Thesis Presentation
• September, 2006

2
Presentation Overview

• Motivation
• Related Work
• LiveRAC Overview and Implementation
• Discussion

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
LiveRAC
3
Problem Domain

• Managed Hosting Services, data center
  operations staff

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
4
Network Devices

• Network device any electronic device that
  connects to a computer network
• Most network devices can be monitored
• Network Operation Centre (NOC) facility for
  monitoring large numbers of network devices

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
5
Monitored Data

• Most data collected from network devices is
  time-series data
• time stamp and value
• Two types of time-series objects collected
• performance metrics
• 10 AUG 2006 95237, CPU, 95
• alarm data
• 10 AUG 2006 95237, MAJOR, HIGH TEMP
• Key difference for visualization
• performance metrics quantitative
• alarms categorical

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
6
Detail Overload
Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
7
RRDTool
Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion

• Database system with statistical graphics
• Monitors individual hosts
• 6-inch view
• Basis for many related applications

http//oss.oetiker.ch/rrdtool/
8
Ganglia

• Cluster monitoring tool, uses RRDtool back-end
• Provides aggregate charts, 100-mile high overview

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
http//ganglia.sourceforge.net/
9
OpenNMS

• Aggregates SNMP data from multiple hosts, uses
  RRDtool back-end
• Alarm management
• 1000-mile high overviews

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
http//opennms.org
10
Visualization Solution Requirements

• Scale to large, dynamic datasets
• thousands of devices
• dozens of data channels
• multiple time scales
• Three levels of activity

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
Back-end DB
LiveRAC
11
Our solution LiveRAC
Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion

• Interactive user-directed exploration of overview
  detail
• Second to sub-second timeframe

12
Our solution LiveRAC

• Reorderable matrix
• rows of network devices
• columns of time-series objects
• Semantic zooming and aggregation for cells
• large cells show time-series charts
• compact representations in reduced areas
• aggregate spatial representation shown in highly
  compressed regions

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
13
Visual Introduction to LiveRAC
14
Research Contributions

• A scalable visualization system called LiveRAC
  using real world data
• Algorithms and code to support dynamic and
  reorderable data elements in accordion drawing
• Infrastructure and algorithms for user-definable
  semantic zoom in accordion drawing

Motivation Overview Problem Domain Devices
Data Domain Tools Requirements Our
Solution Related Work LiveRAC Discussion
15
Information Visualization

• Human visual channel is highest-bandwidth
  perceptual system Norretranders, 1999
• Information visualization field of study whose
  object is to aid cognition through the graphic
  representation of abstract data
• displays relevant information graphically to
  assist in memory tasks
• supports data exploration through direct
  interaction
• assists in pattern finding through the display of
  overview and detail, search, and user-directed
  reordering

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
16
Information Visualization Techniques - A few
examples

• Small multiples

Pre-attentive visual cues
Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
17
Information Visualization Techniques - A few
examples

• Small multiples

Pre-attentive visual cues
Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
18
Time-Series Data

• Extensively explored in information visualization
• Many techniques cluster similar time-series data
  points together, e.g. work by van Wijk et al.
• LiveRAC provides many small-multiples views of
  time-series data

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
J. van Wijk, E. Van Selow. Cluster and calendar
based visualization of time series data. Proc.
IEEE Symposium on Information Visualization, pp 4.
J. Lin, E. Keogh, S. Lonardi, J. Lankford, D.
Nystrom. Visually mining and monitoring massive
time series. SIGKDD 2004, 460-469.
19
Statistical Graphics
Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n

• Statistical graphics
• projection of abstract shapes representing
  observed quantitative data
• in use for centuries in various forms Beniger,
  1978
• used throughout science and industry in commonly
  available tools like Excel

20
Reorderable Visualizations
Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n

• Supports user-directed pattern finding when
  patterns are not known a priori Bertin, 1981
• allows users to group and sort data to identify
  or confirm patterns
• supports an intuition-driven model for
  interacting with data
• LiveRAC adds reordering to accordion drawing

E.Mäkinen, H. Siirtola. Reordering the
Reorderable Matrix as an Algorithmic Problem.
Theory and Application of Diagrams. 2000, 453467.
21
Semantic zooming

• Semantic zooming represents data differently at
  different zoom levels Perlin, 1993
• Optimize representation for available space
• Allow multiple levels of detail

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
R. Bade, S. Schlechtweg, S. Miksch. Connecting
Time-Oriented Data and Information to a Coherent
Interactive Visualization. CHI 2004, pp 105-112.
22
Accordion Drawing

• Information visualization technique
• Stretch-and-squish navigation
• enlarge some areas while retaining surrounding
  context
• Guaranteed visibility
• important landmarks remain visible

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
Munzner, Guimbretiere, Tasiran, Zhou, and Zhang.
TreeJuxtaposer Scalable Tree Comparison using
FocusContext with Guaranteed Visibility.
SIGGRAPH 2003, 453-462.
23
PowerSetViewer

• Dynamic accordion drawing
• insert and remove data at run time
• Limitations
• domain-specific solution
• not user-reorderable

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
Qiang Kong. Visual mining of power sets with
large alphabets. UBC CS, Masters thesis, May
2006.
24
PRISAD

• Introduces the concept of per-frame partitioning
  of data into screen-visible regions
• Reduces n data nodes to p where p is bounded by
  the number of display pixels
• Provides an API for developing accordion drawing
  applications
• Limitation
• static data structures

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
Slack, Hildebrand, and Munzner.Partitioned
Rendering Infrastructure for Scalable Accordion
Drawing (Extended Version) Information
Visualization, 5(2), p. 137-151, 2006.
25
SWIFT

• SWIFT is a set of data storage, aggregation and
  visualization tools that integrate multiple data
  sources Koutsofios, 1999
• Developed at ATT Labs, fully deployed in a
  production role
• Data sources include SNMP, intrusion detection
  systems, Windows system monitors, and custom
  written daemons
• Highly scalable
• Optimized for streaming data

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
26
SWIFT Front-ends

• Existing views
• Geographic views
• Node-link diagrams
• Raw data text
• Limitations
• Cannot compare between large numbers of
  time-series objects
• LiveRAC reorderable matrix visualization for
  SWIFT

Motivation Related Work InfoVis Time-series
Statistical Graphics Reorderable Vis Semantic
Zoom Accordion Drawing SWIFT LiveRAC Discussio
n
Koutsofios, North, Truscott, and Keim.
Visualizing large-scale telecommunication
networks and services.IEEE Visualization 1999,
457-461
27
LiveRAC Architecture
Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
28
LiveRAC Architecture - PRISAD

• Uses and enhances the PRISAD accordion drawing
  API
• PRISAD provides
• well-established scalability
• pixel-bounded rendering performance
• extensive infrastructure
• Contributions
• fully dynamic generic data structures
• add, remove and reorder

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
29
LiveRAC Architecture - SWIFT

• SWIFT back-end provides
• unified interface for multiple collected data
  sources
• temporal aggregation
• Separate rendering and data-service threads allow
  interaction during data retrieval
• Time window can be selected to display historical
  or live data

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
30
Semantic Zooming

• CPU usage at several levels of detail

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
31
Variable LOD Charts

• jGLChartUtil High-performance OpenGL statistical
  graphics library
• Several data representations
• line charts
• scatter charts
• bar charts
• histograms
• sparklines

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
32
Variable LOD Charts

• Optimizes chart representation for best use of
  available space
• scales fonts
• best fit axis labeling
• modifies chart grid
• Representation and size selected by application,
  not library

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
33
Defining Semantic Zoom Levels

• Bundle specification of how to draw cells in a
  column
• defines graphical representation at different
  cell sizes
• can contain single or multiple time-series
  objects
• pre-defined in configuration file
• Generic bundles provide defaults

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
34
Aggregation

• At lowest level of detail, multiple rows are
  aggregated to single visual representation
• convey single useful value for large quantity of
  information
• Aggregated representation is colored box with
  varying saturation
• for alarms, color indicates highest severity
  alarm
• gray indicates metric data
• saturation is a function of density

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
35
LiveRAC Semantic Zooming
Semantic Zooming Video
36
Reordering

• LiveRAC allows rows and columns to be reordered
• rows sorted by device name, or by customer
  identifier and sub-sorted by device name, or
  ordered arbitrarily
• columns ordered arbitrarily, locations specified
  by user
• Required significant extensions to PRISAD to
  provide dynamic data structures

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
37
Reordering Rows and Columns - Visual Example
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
Reordering Video
38
LiveRAC Data Playback

• Playback consists of advancing the time window by
  a configurable duration at regular intervals
• Historical data can be viewed faster than
  real-time
• Current data can be viewed in real-time

Motivation Related Work LiveRAC Architecture
Semantic Zoom Reordering Playback Discussion
39
(No Transcript)
40
Discussion Results

• LiveRAC was deployed using real world data at
  ATT Labs
• LiveRAC proved to be scalable
• interactive frame rates
• 4000 device rows
• dozens of performance/alarm columns

Motivation Related Work LiveRAC Discussion
Results Future Work Conclusion
41
LiveRAC Case Study
Case study video
42
User Feedback

• Demo feedback was positive
• users familiar with the old system were able to
  quickly recognize customers based on familiarity
  with the data
• LiveRAC identified by domain managers as possible
  next-generation tool for data center usage
• users had numerous suggestions for the system, a
  good indication that they were excited by the
  possibilities

Motivation Related Work LiveRAC Discussion
Results Future Work Conclusion
43
Future Work

• Interaction
• field study LiveRAC in real-world environment
• support alternative navigation options
• auto-expanding search region
• hot-keying groups of devices or metrics
• expand data representation library
• Data processing
• computational correlation of alarm and metric
  data
• Performance
• lazy evaluation during reordering

Motivation Related Work LiveRAC Discussion
Results Future Work Conclusion
44
Conclusion

• Contributions
• working system for interactive visualization of
  large real-world time-series data sets
• algorithms for reorderable accordion drawing
• infrastructure for semantic zoom in accordion
  drawing

Motivation Related Work LiveRAC Discussion
Results Future Work Conclusion
45
(No Transcript)
46
(No Transcript)
47
(No Transcript)
48
Targeted User Activities

• Overview
• Situational awareness
• Critical alarm notification
• Detail
• Incident investigation
• Capacity planning
• Trend analysis

Motivation Related Work LiveRAC Architecture
Visual Encoding Aggregation Semantic Zoom
Reordering Playback Discussion Future
Work Conclusion
49
Monitoring Large-scale Systems A Difficult
Problem

• Many disparate data sources
• Different platforms
• Different protocols
• Different services
• Lack of integrated solutions
• Lack of context in standard tools
• Shortage of integrated visualization solutions

Motivation Related Work LiveRAC Architecture
Visual Encoding Aggregation Semantic Zoom
Reordering Playback Discussion Future
Work Conclusion
50
Semantic Zooming in LiveRAC

• LiveRAC semantic zooming
• provides an area-aware graphical representation
• modifies the graphic to best fit the available
  space
• does not increase or decrease polygon count, or
  scale a graphic linearly, but changes graphic
  attributes, and may change the graphic entirely

Motivation Related Work LiveRAC Architecture
Visual Encoding Aggregation Semantic Zoom
Reordering Playback Discussion Future
Work Conclusion
51
SWIFT Limitations

• Visualizations for SWIFT only provide node-link
  and geographic views
• These views are good for mapping physical
  relationships

Motivation Related Work LiveRAC Architecture
Visual Encoding Aggregation Semantic Zoom
Reordering Playback Discussion Future
Work Conclusion
52
SWIFT Architecture
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
53
LiveRAC handles rows and columns differently
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion

• Large number of rows in a typical data view
  (thousands)
• Comparatively small number of columns (dozens)

54
Performance Requirements

• LiveRAC must maintain interactive frame rates
  while modifications to the grid are taking place
• We need to draw guaranteed visible zones first to
  provide landmarks
• The system needs to scale to thousands of
  devices, and tens of categories of monitored
  alarms metrics
• The system must support a large number of data
  points for alarms metrics

Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
55
Split Line Performance
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
56
Charting Performance

• System 3Ghz Pentium-IV
• Chart 3 data series, 100 points each series
• First draw 50ms
• Subsequent redraws after modification lt5ms
• Redraw from cached OpenGL display list lt1ms

Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
57
Statistical Graphics
Line chart
Bar chart
Scatter plot
Pie chart
58
LiveRAC Visual Encoding

• Matrix view
• rows of devices
• columns of metrics and alarms time-series
  objects
• a cell contains a representation for a set of
  values of any time-series object
• at highest density, cells are colored boxes
• at lower densities cells can contain text, or
  graphical representations

Motivation Related Work LiveRAC Architecture
Visual Encoding Aggregation Semantic Zoom
Reordering Playback Discussion Future
Work Conclusion
59
Accordion Drawing Split Line Structure
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Future
Work Conclusion
Split lines
Slack. Partitioned Rendering Infrastructure For
Stable Accordion NavigationM.Sc. Thesis
60
Accordion Drawing Split Line Structure
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Future
Work Conclusion
Slack. Partitioned Rendering Infrastructure For
Stable Accordion NavigationM.Sc. Thesis
61
Static Split Lines

• Previous accordion drawing implementations used
  static, ordered lists of split lines
• Insert, remove and reorder operations were O(n)

Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
62
Dynamic Split Lines

• Dynamic split lines are required for maintaining
  interactive frame rates while adding/removing
  rows or columns to the matrix
• Client-server streaming architecture implies that
  new devices, alarms and metrics will be a common
  occurrence
• Dynamic split lines support reordering of data
• In the future, direct user modification of the
  data set can be supported

Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
63
Dynamic Split Line Requirements

• Requirements
• Worst case logarithmic insert and remove
  operations
• O(log n) worst case path to any node from the
  root
• Linear scalability in memory usage
• Support for arbitrary ordering with enumeration
• Can we use a red-black tree?
• Red-black trees address the first three
  requirements
• How can we avoid re-enumeration of keys if we
  allow nodes to be manipulated arbitrarily?

Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
64
Dynamic Split Lines Solution
Solution Maintain sub-tree sizes
Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
Find node index 3
3 lt (31), descend left
3 gt (1 1), descend right
found index 3
1 2 3 4
5
65
Reordering Rows and Columns Implementation

• For rows, we can swap pointers from split lines
  to devices without changing the split line
  structure
• O(1)
• Preserves layout topology
• For columns, we use a global ordering list,
  mapping a bundle/metric/alarm name to an index
  number, changes to this list affects rendering of
  all columns
• A hash map is maintained to back map column name
  to index numbers for O(1) lookups during
  rendering, this must also be fixed during a
  reorder
• Swapping columns is O(c), where c is the number
  of columns, typically lt 100

Motivation Related Work LiveRAC Overview
Architecture Dynamic Structure Reordering
Aggregation Semantic Zoom Performance Discussio
n Future Work Conclusion
66
(No Transcript)