SLIDE 1 Pentaho Business Analytics Evolves
Pedro Alves Pentaho SVP Community & Product Designer, Hitachi Vantara
SLIDE 2 Agenda
- Pentaho Vision
- High Level Roadmap
- Pentaho Analytics – Yesterday
- Today: The Beginning of a Journey
SLIDE 3
Pentaho Vision
SLIDE 4
A product vision must be a combination of 3 elements
SLIDE 5 What the market wants What we have What we want
SLIDE 6
Top down approach: Strive for something the users desire
SLIDE 7
Usability first
SLIDE 8
We won’t improve the product usability if we think about technology
SLIDE 9
We have to think differently
SLIDE 10
So instead of thinking about Datasources, APIs, ETL, Models, Dashboards, Architecture…
SLIDE 11
Lets think about what the users want
SLIDE 12
They want information
SLIDE 13 Pentaho Vision – Making it simple
Pentaho Proprietary and Confidential
SLIDE 14 Pentaho Proprietary and Confidential
Pentaho Vision: The story of an insight
SLIDE 15 Pentaho Proprietary and Confidential
Data Preparation
No more separation between DI and Analytics
SLIDE 16 Pentaho Proprietary and Confidential
Data Preparation
Analytics must be available anywhere
SLIDE 17 Pentaho Proprietary and Confidential
Data Preparation
And collaboration is fundamental
SLIDE 18 Pentaho Proprietary and Confidential
Collaboration
This insight then comes to life
SLIDE 19 Pentaho Proprietary and Confidential
Sharing and consumption
A library of insights, In a governed, scalable and collaborative universe
SLIDE 20 Pentaho Proprietary and Confidential
Different outputs
DET Self Explore Visualizations Report Self Service Dashboard CTools Dashboard
And consumed any way needed
SLIDE 21
How are we executing on this vision?
SLIDE 22
High-level Roadmap
SLIDE 23 Pentaho Proprietary and Confidential
Message: A Single Flow with Analytics Anywhere
SLIDE 24 Big Data Processing
Data Explorer Visual Data Prep Streamlined Modeling Data Science Extensions
Visual Data Experience
Spark Engine Adaptive Execution Streaming Machine Learning
(Big) Data Processing
Pentaho Scale-Out Cloud Deployment Lumada Integrated
Enterprise Platform
From Vision to Roadmap
SLIDE 25 Big Data Processing
Data Explorer Visual Data Prep Streamlined Modeling Data Science Extensions
Visual Data Experience
Pentaho Proprietary and Confidential
Visual Data Experience
SLIDE 26 Pentaho Proprietary and Confidential
Spark Engine Adaptive Execution Streaming Machine Learning
(Big) Data Processing
(Big) Data Processing
SLIDE 27 Pentaho Proprietary and Confidential
Pentaho Scale-Out Cloud Deployment Lumada Integrated
Enterprise Platform
Enterprise Platform
SLIDE 28
Pentaho Analytics: Yesterday
SLIDE 29
Analytics have always been relegated to the Pentaho Server
SLIDE 30
SLIDE 31
On a tool based approach
SLIDE 32
SLIDE 33
That we chose in the first place
SLIDE 34 Big Data Processing Analyzer Interactive Reporting Pentaho Dashboards
Pentaho Proprietary and Confidential
CTools Dashboards
Pentaho Server Analytics Tools
SLIDE 35
With different and sometimes overlapping capabilities
SLIDE 36 Big Data Processing
Ad-hoc Pivot View Visualizations Inline Modelling Several functions Ad-hoc Row Level analysis Metadata Based Report generation
Interactive Reporting
Self Service Template based Limited set of components Easy to use
Pentaho Dashboards
Pentaho Proprietary and Confidential
Pre-canned Dashboard Embedded capabilities Allows the Art of the possible Not easy to use
CTools Dashboards Analyzer
Pentaho Server Analytics Tools
SLIDE 37 Today: The Beginning
SLIDE 38
In order to fix usability, we had to take a step back
SLIDE 39
The first ask can’t be:
What tool do you want to use?
SLIDE 40
Analyzer is based on the Mondrian engine
SLIDE 41
Visualizations were tied to it and not reusable
SLIDE 42
Interactive Reporting is based on the Pentaho Metadata engine
SLIDE 43
Both require separate modelling exercises
SLIDE 44
And don’t share common functionality
SLIDE 45
Don’t get me wrong…
SLIDE 46
They are both excellent tools for the use cases they were created for!
SLIDE 47
But we pay the price for maintaining two different applications
SLIDE 48
And users pay the price for the lack of coherency
SLIDE 49
We want to fix it
SLIDE 50
We want to fix it.
And we will!
SLIDE 51
So we’re aiming for 2 objectives:
SLIDE 52
- 1. A unified and coherent way to explore data
SLIDE 53
- 2. That can be used anywhere in the cycle
SLIDE 54
Internal Project Codename: DET
SLIDE 55
We envisioned a tool that abstracted all those concepts
SLIDE 56
And could, in time, provide the features each existing tool provides
SLIDE 57
And be the place to add new capabilities
SLIDE 58
Such tool wouldn’t be built from scratch…
SLIDE 59
Instead, would leverage existing technology
SLIDE 60
But hiding it from the end user
SLIDE 61
They don’t care about models, engines, any of that crap…
SLIDE 62
Now…
SLIDE 63
Some of you may not know…
SLIDE 64
But a first version of this tool already exists since 7.0
SLIDE 65
If you haven’t seen it, you’re looking in the wrong place
SLIDE 66
It’s not here…
SLIDE 67
It’s here
SLIDE 68
We’re building something that allows users to explore any data
SLIDE 69
No matter where it comes from
SLIDE 70
And giving them the freedom to look at it any way they choose from
SLIDE 71
At a row level…
SLIDE 72
SLIDE 73
As a Pivot table
SLIDE 74
SLIDE 75
Or any other visualization
SLIDE 76
SLIDE 77
Since 7.0, we’ve been working on several threads:
SLIDE 78 Increased functionality
Drilldown Filters Actions (Keep only / Exclude / etc)
SLIDE 79 Extensibility and coherency
Viz Api 3.0 Support for Analyzer and CTools
SLIDE 80 Performance improvements
Reducing start-up time Optimizing resources Handling more data
SLIDE 81
And going forward we’ll keep working on many more
SLIDE 82
More functionality
SLIDE 83
Inline modeling capabilities
SLIDE 84
Profiling capabilities
SLIDE 85 Support for bigger and more complex datasets
But thinking about this differently….
SLIDE 86 Write-back capabilities (Wrangling)
The real PDI thin client strategy, a wysiwyg ETL layer
SLIDE 87 Ability to plug in your own data
Always in a governed way
SLIDE 88
Available everywhere: PDI, Pentaho Server…
SLIDE 89
Or embedded in your own application
SLIDE 90
Pentaho Analytics: Tomorrow
SLIDE 91
Going back to the original vision
SLIDE 92
We want to remove the focus from the tool
SLIDE 93
And put the focus on the information
SLIDE 94
The insights we get from them
SLIDE 95 And let everything else follow…
Visualizations Self service ”dashboards” Printed reports Custom Dashboards Blending on the glass
SLIDE 96
Allowing us to implement the original vision
SLIDE 97 Pentaho Proprietary and Confidential
DET Self Explore Visualizations Report Self Service Dashboard CTools Dashboard
Pentaho Analytics
SLIDE 98