This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>


To have a better experience, please upgrade your IE browser.



  • Agile

    Offline and online data processing with response times down to the subsecond and even millisecond range.

    Secondary development APIs and environments enable organizations to accelerate their specific rollouts by 3-fold to 10-fold their original speeds.

    User-friendly O&M system simplifies system maintenance.

    Agile business migration capabilities fully satisfy SQL standards; organizations can smoothly migrate conventional services to the Big Data platform without having to modify code.

  • Smart

    Full data analysis capabilities drill down with pinpoint accuracy into millions of dimensions.

    Supports high-performance analysis and query, as well as multi-dimensional comparisons.

  • Trustworthy

    HA of all components.

    First Big Data software that fully complies with China’s information security standards for the finance industry.

    Enhanced reliability and security.

Solution Architecture

Huawei’s Big Data Solution consists of two products: FusionInsight HD and FusionInsight LibrA. FusionInsight HD is a Hadoop enterprise edition containing many components: HDFS, Yarn, HBase, Spark, MapReduce, Flink, Storm, Elk, Solr, Kafka, Loader, Flume, and so on. FusionInsight LibrA is a massively parallel processing database that features elastic scalability, excellent performance, rock-solid reliability, and superior cost-effectiveness.

  • HDFS

    Provides data access with high throughput; can process large-scale data sets.


    As the resource management system of Hadoop 2.0, Yarn implements resource management and scheduling for applications.


    An in-memory distributed computing framework.


    Provides standard SQL engine and enables conventional applications to be smoothly migrated to the Big Data platform.


    A distributed computing engine supporting massive offline batch processing.


    A unified computing framework for batch and stream processing and stream processing. At its core is a stream processing engine that supports data distribution and parallel computing.


    A distributed, reliable, and fault-tolerant real-time stream data processing system. It provides SQL-like query languages (StreamCQL).


    An independent, enterprise-class application search server based on Apache Lucene.


    A distributed, partitioned message release-subscription system with multiple copies.


    Exchanges data and files between FusionInsight, relational databases, and file systems.


    A column-oriented distributed storage system suitable for mass unstructured or semi-structured data that provides high availability, performance, and scalability. HBase supports real-time data read and write.


    A distributed mass log collection, aggregation, and transmission system that provides high availability and reliability.


    A massively parallel processing database that features elastic scalability, excellent performance, rock-solid reliability, and superior cost-effectiveness. It can replace conventional data warehouse systems and adds new levels of efficiency to decision-making.

Application Scenarios

  • Converged Data Warehouse
    • Conventional data warehouses are cumbersome, costly, and difficult to scale out. Complicating matters, much of the hardware is incompatible, and many of the in-place platforms cannot cope with the ever-growing data analysis and decision-making requirements. The Huawei converged data warehouse features horizontal expansion, HA of all components, hybrid row-column storage, high-speed data query and analysis, compatibility with conventional SQL applications, and smooth application migration. Customers in the carrier, finance, public security, and many other sectors benefit from the enhanced agility of the Huawei solution.

  • Offline Processing
    • Massive sets of data are analyzed and processed, and the results are provided for later use. Offline processing jobs do not have intense requirements on processing time, but the data to be processed is in diverse formats and often reaches petabyte-scale. Offline processing frequently involves multiple MapReduce, Spark, Hive, and Spark SQL jobs and applies to data preprocessing and offline analysis in the finance, carrier, public security, and many other sectors.

  • Interactive Query
    • Interactive analysis and query have demanding requirements on response time and involve massive sets of data. Data used for interactive query is usually pre-processed based on data models suitable to the task. Interactive query includes precise query, ad hoc query, theme analysis, and other tasks.

  • Real-Time Processing
    • Data is collected and analyzed in real time, and analysis results are provided in a timely manner. Capable of rapidly processing data from diverse sources at high throughput, real-time processing applies to anti-fraud monitoring, real-time security surveillance, and other scenarios.

Success Stories


How to Import and Analyze Data Based on FusionInsight HD

Cooperation and Ecosystem

Open Source Ecosystem
  • No. 2

    Huawei ranked second among all global contributors in the OpenStack community.

  • No. 6

    Huawei ranked first in overall contributions among Chinese contributors and sixth among all global contributors in CNCF.

  • No. 3

    Huawei ranked third among all global contributors in the Hadoop community.

  • No. 4

    Huawei ranked fourth among all global contributors in the Spark community.

Huawei OpenStack Introduction