en
0.25
0.5
0.75
1.25
1.5
1.75
2
Scale-out Beyond MapReduce
Published on Sep 27, 201320637 Views
The amount of data being collected is growing at a staggering pace. The default is to capture and store any and all data, in anticipation of potential future strategic value, and vast amounts of data
Related categories
Chapter list
Scale-out Beyond MapReduce00:00
Outline00:44
Cloud Information Services Lab (CISL)01:35
Big Data - What’s the big deal?02:23
What’s New?02:30
Challenges02:56
Web of Concepts04:01
An Example04:58
Content Optimization06:29
CORE Dashboard: Segment Heat Map07:48
Kinect09:08
Kinect-based Full Body Gait Analysis09:58
Connected devices will soon be EVERYWHERE10:17
HomeOS: Another Instance of IoT10:43
Big Data - Build it, they’re here already!12:29
One Slide MapReducePrimer - 113:02
One Slide MapReducePrimer - 213:27
One Slide MapReducePrimer - 313:31
One Slide MapReducePrimer - 413:33
One Slide MapReducePrimer - 513:48
One Slide MapReducePrimer - 613:50
One Slide MapReducePrimer - 713:53
One Slide MapReducePrimer - 814:08
One Slide MapReducePrimer - 914:11
One Slide MapReducePrimer - 1014:12
One Slide MapReducePrimer - 1114:15
The Digital Shoebox15:46
Microsoft21:00
A Common Vision22:10
Challenges22:53
How Far Away is Data?24:02
Compute Fabric: YARN24:09
Making YARN Easier to Use: REEF26:32
Digital Shoebox Architecture27:42
Motivation: Machine Learning Workflow29:19
Example: User Activity Modeling30:10
Feature and Target Windows30:54
User Modeling Pipeline31:26
Example Formation: SQL at Scale31:52
Learning a Language Classifier33:08
Scaling Model Building (30,000ft)34:02
The Challenge - 135:18
The Challenge - 235:51
Machine Learning Workflow37:08
Take-Away38:06
REEF: Retainable Evaluator Execution Framework38:24
What have we built on top of REEF?38:35
REEF in the Stack39:04
REEF: Computation and Data Management39:25
Running Example: Distributed Shell40:51
The REEF Control Flow - 141:02
The REEF Control Flow - 242:04
The REEF Control Flow - 342:16
The REEF Control Flow - 442:53
The REEF Control Flow - 542:56
The REEF Control Flow - 643:02
The REEF Control Flow - 743:03
The REEF Control Flow - 844:13
The REEF Control Flow - 944:36
The REEF Control Flow - 1045:07
The REEF Control Flow - 1145:13
REEF Control Flow: Summary45:17
Learning in REEF45:46
The Task: Learn a Regression Model45:51
Linear Models45:57
The Learning Algorithm: Batch Gradient Descent46:00
How It Maps to REEF46:20
How It Maps to REEF: Control Flow47:03
Contrast: Hadoop MapReduce47:34
Data Management Services47:58
Contributing to Apache48:12
There is More, But Not Today48:54
Conclusions49:02