Big data presentationandoverview_of_couchbase

AGENDA
 B U S I N E S S S C E N A R I O S - B I G D A T A
 Technologies solving business challenges
 Big data Tools – Solving Business Cases
 E N G A G E M E N T D A T A B A S E V S T R A N S A C T I O N A L
 W H Y C O U C H B A S E
 Couchbase Key Capabilities
 Couchbase Unique Value Proposition
 Fourth dimensions of Big Data
 Framework – Layered Approach
 Tools & Technologies
 Q U E S T I O N S

3
Business Scenarios – Need of Big Data
Big
Data
Actionable
Insights with
Right Team
Customer
Experience
with
Connected
Device
Data
Governance &
Compliance
Provide
Insights for
Humans and
AI Tools
Choose right
technology
and right
platform
Leverage proper
model to align
with right product
for right
customers

4
Technologies solving business challenges
Batch Use cases
Analytic ( )
Real time Use cases
Engagement ( )
 Customer segmentation,
 Associations of products,
 Customer churn analysis,
 Cart abandonment
 Fraudulent transactions,
 Credit card application validation,
 Customer churn analysis,
 Customer experience analytics
 Personalization
Hadoop features :
 To process large volume of data
 Take seconds for processing
 Variable based on the load
 Disk-Centric
 Can handle Petabytes of data
 Batch processing and Adhoc
Couchbase feature:
 Real time and Interactive access to data
 Very low latency < 1ms ops
 Predictable performance
 Memory-Centric
 Can handles 10s of Terabytes
 Online and Adhoc

5
Big data Tools – Solving Business Cases
Spark Use Cases Kafka Use Cases Storm Use Cases Flink Use Cases
 Real-time recommendations
 Component failure detection
 Network intruder detection
 Fraud detection
 Web Activity tracking
 Log Aggregations
 Event sourcing
 RSS Feed stream analysis
 Social media analytics
 Real time trade-in analytics
 Malfunction detection
 Log & Metric analysis
Fraud detection
Ana moly detection
Rule based alerting
Real-time search index building
Spark features:
 Spark uses Dstream for
streaming data RDD (Resilient
distributed dataset) for batch
data.
 Cache dataset mainly used for
ML
 Persist RDD
 Micro-Batch processing
Kafka features:
 Distributed messaging system
 Acts as a Message Broker
 Persist data in File system such
as EXT4 OR XFS
 Small-batch processing
 Zookeeper (scheduler)
dependent
Storm features:
 Real-time messaging
 Stream processing
 Data is not persisted
 Micro-batch processing
Flink features:
 Built exclusively for streaming
 Data is not persisted.
 Use checkpoints on streaming
data to finite sets

6
Characteristics of RDBMS Versus Engagement (NOSQL) Databases
RDBMS NOSQL Databases
Moderate velocity of data High velocity of data (devices, sensor..etc.)
Data coming in from one/few locations Data coming from many locations
Primary structured data Structured, with semi-unstructured
Complex/Nested transactions Simple transactions
Protect uptime via failover/log shipping Protect uptime via architecture
High Availability Continuous Availability
Deploy app central location/one server Deploy app everywhere/many servers
Primary concerns: scale reads Scale reads/writes
Scale up more users/data Scale out for more users/data
Maintain data volumes with Purge High data volumes; retains forever.

Couchbase - Key Capabilities
 ANSI Join is a
standardized join syntax
widely used in relational
databases.
 Eventing Service
enables server-side
functions using the
familiar EventCondition-
Action mode
Automate multi-cloud
deployment and
management with
Kubernetes.
▪ Auto failover for common
disk failures without
operator intervention
▪ Compliance with
enhanced certificate
authentication, auditing
support, and log redaction
on user sensitive data
 Ease of deployment
 Gain app insights
Large scale N1QL
deployments with hash
joins, faster aggregates, and
index partitioning.
▪ Orders of magnitude faster
N1QL queries with hash
joins and aggregate
optimization
▪ End-to-end data
compression from client to
server
 Find answer faster
 Scale more easily
 Improve TCO
Agility Manageability Performance

End-to-end Compression GSI INDEX PARTITIONING
Couchbase Unique Value Proposition
AGGREGATE PUSHDOWN
INDEX BUILD
TIME 50%
INDEX SIZE
-70%
FTS SCORCH INDEXING (DP)

Strategy :The fourth dimension of Big Data:
Volume Velocity Veracity*
9
Variety
Data at Rest
Terabytes to exabytes
of existing data to
process
Data in Motion
Streaming data,
milliseconds to
seconds to respond
Data in Many
Forms
Structured,
unstructured, text,
multimedia
Data in Doubt
Uncertainty due to data
inconsistency
& incompleteness,
ambiguities, latency,
deception, model
approximations

Framework – Layered Approach
Collection Layer
Data collection, registration and Profile
Storage & Consolidation Layer
Storage solution for structured/
Unstructured data
Process Layer
Data mining data set
Point of Sale data set
Reporting data set
Analytical Layer
Miscellaneous ways in
which data is consumed to
derive value
• Predictive analytics
• Ad-hoc analytics and
optimization.
• Segmentation, Profiling.
• Customer tastes, affinities
an segments
• Recommendations and
context engines.
Monetization and
Experience
• Personalization and
experiences.
• Brand Marketing
• Channel and Operation
Optimization
Payable Omniture Other

Tools and Technologies
Ingest/Propagate
Apache Flume, Apache Kafka, Apache Scoop, Spark, Storm, Flink
Implement, Search, Analyze
Apache MapReduce, Apache Solr, Apache Blur and Lucene
Apache Hama, Apache Giraph, Cloudera Impala, Apache Storm, Apache HBase
Persist
File System: Apache HDFS, MapR Distributed File System
Serialization: Apache Avro
DBMS: Apache Cassandra, Couchbase ,Apache Hbase, Apache Accumulo, MongoDB
Monitor, Administer, Manage
Apache Ambari, Apache Oozie, Apache Zoopeper, Cloudera Enterprise Manager
Advanced Analytics and Machine Learning
Apache Drill, Apache Mahout, Datameer, Rhadoop, IBM SPSS, SAS, MatLab, R, D3, Python

Big data presentationandoverview_of_couchbase

More Related Content

What's hot (18)

Similar to Big data presentationandoverview_of_couchbase (20)

Recently uploaded (20)

Big data presentationandoverview_of_couchbase

Editor's Notes