FusionInsight Big Data Platform

FusionInsight provides a comprehensive Big Data software platform for batch and real-time analytics using open-source Hadoop and Spark technologies.

The system leverages HDFS, HBase, MapReduce, and YARN/Zookeeper for Hadoop clustering, along with Apache Spark for faster real-time analytics and interactive queries. Solr adds powerful full-text searching of rich text documents (Word and PDF files), and rich APIs and development tools let you customize the system for specialized data analysis.

Extract big value from Big Data faster and easier with Huawei’s enterprise-class FusionInsight data analysis platform.

Features

FusionInsight brings Big Data Hadoop and Spark technologies together in an integrated, enterprise-class software platform for faster data analysis and better decision-making

  • Optimized for agility: Comprehensive, fully featured Big Data analytics platform with open architecture and APIs supporting batch processing, micro-batch processing, and real-time processing for flexible analysis and integration with enterprise data processing
  • Smart: Over one million dimensions in data modeling enable deep insights into user behaviors, helping enterprises to quickly make decisions and respond to market and business opportunities
  • Trustworthy: Reliable, high-performance data processing with the reliability, stability, and security expected in enterprise-class applications and mission-critical financial systems

Specifications

Performance Specifications

Component  Processing Metrics and Response Times System Environment
Parallel Computing Engine (MapReduce)
  • WordCount: Average processing capability of a node: 8 GB/minute
  • Terasort: Average processing capability of a node: 6 GB/minute
 Cluster scale: 12 nodes

Typical node configuration:
CPU: 2 x E5-2650
Memory: 128 GB
Disk: SATA
Parallel Computing Engine (Spark)
  • WordCount: Average processing capability of a node: 27 GB/minute
  • Terasort: Average processing capability of a node: 6 GB/minute
Hive
  • Processing capability — HiveAggregation: Average processing capability of a node: 8 GB/minute
  • Processing capability — HiveJoin: Average processing capability of a node: 2 GB/minute
HBase
  • 100% random read: 30,000 records/s (Average number of records read by each node. The size of a record is 1 kB. The response time is less than 50 ms.)
  • 100% random write: 37,000 records/s (Average number of records written by each node. The size of a record is 1 kB. The response time is less than 50 ms.)
  • Sequential scan: 10,000 records/s (Average number of records scanned by each node. The size of a record is 1 kB. The response time is less than 50 ms.)

Algunas de nuestras REPRESENTACIONES