Rebaca’s Log Analysis service can process any logs and machine data. One can diagnose and visualize critical logs and identify recurring errors. Our intuitive UI allows domain experts to create rules and generate recommendations. Our Kafka-Spark-Cassandra framework can handle streaming & batch data from disparate data sources. It can ingest high volumes of data at an exponential speed to avoid system latencies. One can generate actionable business insights at ease.

RCA Log Analysis

Rebaca offers proactive BI analytics on application logs, infrastructure logs, and systems data. We have deep expertise on Telecom, Video and underlying operating systems. Domain knowledge enables us to correlate and detect anomalies across the system stack. We have the expertise to associate and index any application logs and machine data. We use adaptive machine learning techniques to provide advanced analytics and business intelligence.

Here are some of the logs we process: error logs, syslogs, transaction & event logs, message logs, application logs, DB access logs, all infra logs and more machine data


Using Big Data Solutions to provide business insights and predictive captabilities

Features and Specification

Real-time Multi log ingestion

Logstash pipelining tool, Fluend

Read More

Log indexing, processing & correlation

Elasticsearch DB, Solr

Read More

Log analytics/ Visualization & Recommendation/Prescriptive Analytics

Kibana/Tableau, D3.js, angular.js

Read More

Machine Learning

Neural Network, Support Vector Machine

Read More


1) Mobile Network:
Operation log analysis of to determine underlying error condition and manage customer SLA. We ingest logs generated by Packet Core network components and system KPI. Domain knowledge helps correlate data and utilize machine learning techniques to identify anomalies. We create rules against error and anomalies to provide recommendations.
• Use of packages across a segment of a network
• Seasonal behavior of a subscriber base and individual subscriber
• Usage pattern based on time of day
• Production error due to incorrect system configuration or sizing
• Error caused due to faulty system hardware components
• Error caused due to incorrect behavior of 3rd party components
• Remedial recommendations against error conditions


2) Video Solution:
Logs from different components of Video Delivery Network are analyzed to derive root cause behind error conditions. Our years of deep domain knowledge on every video subsystem enable us to provide automatic log parsing techniques such that any structure or unstructured logs can be ingested and indexed properly for viewing. Rules can be created using a intuitive UI to correlate log messages along with system KPIs. Recommendations and remedial techniques can be associated with error conditions for faster resolution.
• Customer viewing habits based on content type and category
• Content viewing pattern across a geographical area
• Seasonal viewing habits
• Production error due to incorrect system configuration or sizing
• Error caused due to faulty system hardware components
• Error cause due to DRM license interoperability issues
• Error caused due to incorrect behavior of 3rd party components


3) 360-degree Customer View:
A complete analysis of customer’s content viewing profile. Aggregation of data from the various touch points viz mobile apps, web portals. The application targets to increase customer loyalty and satisfaction by understanding habits
• Customers activity in app/portal round the clock
• Predictive analysis based on historical time series customers purchase propensity
• Customers review on sites like facebook and twitter
• Customer sentiment from both internal and external sources
• Customer retention analytics by analyzing customers active & inactive period of stay.


4) Social Feed:
Social media analytics by getting a social feed from blogs and social media websites. Customer sentiment, marketing, and customer service are the common use of social media analytics.
We work on the following social feed data:
• Textual data (such as Tweets and comments/posts)
• Network data (such as Facebook Friendship Network, and Twitter follow-following network)
• Actions (such as likes, shares, views)
• Hashtag/Hyperlinks (e.g., hyperlinks embedded within text)
• Mobile data (e.g., mobile application data)
• Location data
• Search engines data


5) Analytics on on-premise network & security data from UTM:
We provide SPAM analysis to ensure that emails always reach their target. By doing accurate spam analysis organization can automate email authentication. Users can send emails from any domain and protect from spam sites.
Spam anomalies analyze the following stats:
• Viz. Email header: where the spam starts, spam domain
• Return-path, Received tags, X-IP tag, X-UIDL of spam mail etc.
• Spam mail stats from penalized/banned sites/portals
• Referrer spam stats
• Disavowing spam link stats
• Security threats/attacks analysis