Reviews
Now Reading
Jubatus
1
Review

Jubatus

Overview
Synopsis

Jubatus is the first open source platform for online distributed machine learning on the data streams of Big Data.

Category

Data Mining Software Free

Features

•Scalable
•Real-Time
•Difference from Hadoop and Mahout
•Deep-Analysis

License

Open Source

Price

Free

Pricing

Subscription

Free Trial

Available

Users Size

Small (<50 employees), Medium (50 to 1000 Enterprise (>1001 employees)

Website
Company

Jubatus

What is best?

•Scalable
•Real-Time
•Difference from Hadoop and Mahout

What are the benefits?

•Scalable
•Real-Time
•Deep-Analysis
•Designed for clusters of commodity
•Supports basic tasks

PAT Rating™
Editor Rating
Aggregated User Rating
Rate Here
Ease of use
7.6
8.2
Features & Functionality
7.6
8.1
Advanced Features
7.6
8.3
Integration
7.6
Performance
7.6
Customer Support
7.6
8.2
Implementation
Renew & Recommend
Bottom Line

Jubatus uses a loose model sharing architecture for efficient training and sharing of machine learning models, by defining three fundamental operations; Update, Mix, and Analyze, in a similar way with the Map and Reduce operations in Hadoop.

7.6
Editor Rating
8.2
Aggregated User Rating
2 ratings
You have rated this

Jubatus supports basic tasks including classification, regression, clustering, nearest neighbor, outlier detection, and recommendation. Jubatus is the first open source platform for online distributed machine learning on the data streams of Big Data. Jubatus uses a loose model sharing architecture for efficient training and sharing of machine learning models, by defining three fundamental operations.

Update, Mix, and Analyze, in a similar way with the Map and Reduce operations in Hadoop. In addition, Jubatus supports scalable machine learning processing. It can handle 100000 or more data per second using commodity hardware clusters. It is designed for clusters of commodity, shared-nothing hardware.

Moreover, Jubatus updates a model instantaneously just after receiving a data, and it analyze the data instantaneously. Jubatus supports most tasks for deep analysis, anomaly detection, cluster analysis, simple statistics, and graph analysis. Jubatus also processes data in online manner, and achieve high throughput and low latency. To achieve these features together, Jubatus uses a unique loosely model synchronization for scale out and fast model sharing in distributed environments.

Jubatus processes all data in memory, and focus on operations for data analysis. Users can also use jubakit which is a Python module to access Jubatus features easily. jubakit can be used in conjunction with scikit-learn so that users can use powerful features like cross validation and model evaluation.

Jubatus officially support Red Hat Enterprise Linux (RHEL) 6.2 or later (64-bit) and Ubuntu Server 14.04 LTS / 16.04 LTS (64-bit). On supported systems, users can install all components of Jubatus using binary packages. Other Linux distributions (including 32-bit) and Mac OS X are experimentally supported.

Filter reviews
User Ratings





User Company size



User role





User industry





1 Reviews
  • Bruno Shofner
    September 14, 2017 at 11:28 am

    Manage bigger data, apply deep analytics, and process in real-time

    Company size

    Enterprise (>1001)

    User Role

    IT Support

    User Industry

    Health care

    Rating
    Ease of use8.2

    Features & Functionality8.1

    Advanced Features8.3

    Customer Support8.2

    ADDITIONAL INFORMATION
    Jubatus is an open source online machine learning and distributed computing framework that is developed to manage bigger data, apply deep analytics, and process in real-time. The framework updates a model in real-time after receiving data and analyzing it. Jubatus is designed for clusters of commodity, shared-nothing hardware and supports scalable machine learning processing. The open source platform also supports basic tasks including classification, regression, and recommendation. Jubatus provides deep-analysis including classification, regression, nearest neighbor, recommendation, anomaly detection, clustering, cluster analysis, simple statistics, and graph analysis. The system can handle 100000 or more data per second using commodity hardware clusters and is designed for clusters of commodity, shared-nothing hardware. Jubatus uses a loose model sharing architecture defining three fundamental operations to update, mix, and analyze. Jubatus processes all data in memory with a focus on operations for data analysis using a unique model synchronization for scale out and fast model sharing in distributed environments.

Ease of use
Features & Functionality
Advanced Features
Integration
Performance
Customer Support
Implementation
Renew & Recommend

What's your reaction?
Love It
0%
Very Good
0%
INTERESTED
0%
COOL
0%
NOT BAD
0%
WHAT !
0%
HATE IT
0%