Process for Using GaussDB(DWS) Cluster

Accessing a Cluster

GaussDB(DWS) provides a web-based management console and HTTPS-compliant APIs for you to manage data warehouse clusters.

Accessing the Database in a Cluster

GaussDB(DWS) supports database access using the following methods:

GaussDB(DWS) clients

Access the cluster database using GaussDB(DWS) clients. For details, see Using the gsql Client to Connect to a ClusterUsing the Data Studio GUI Client to Connect to a Cluster, and Using DAS to Connect to a Cluster.

JDBC and ODBC API calling

You can call standard APIs, such as JDBC and ODBC, to access databases in clusters.

For details, see Using a JDBC Driver to Connect to a Database and Using an ODBC Driver to Connect to a Database.

psycopg2 and PyGreSQL drivers

After creating a data warehouse cluster, you can use the third-party function library psycopg2 or PyGreSQL to connect to the cluster, and use Python to access GaussDB(DWS) and perform various operations on data tables. For details, see Using the Third-Party Function Library psycopg2 of Python to Connect to a Cluster and Using the Third-Party Function Library PyGreSQL of Python to Connect to a Cluster.

End-to-End Data Analysis Process

GaussDB(DWS) has been seamlessly integrated with other services on Huawei Cloud, helping you rapidly deploy end-to-end data analysis solutions.

The following figure shows the end-to-end data analysis process. Services in use during each process are also displayed.

Activities & Promotions Activities & Promotions

Free Packages

Register now to take advantage of these incredible deals

Hot
Try Now
Huawei Cloud Blossom Program

Help your business bloom with cloud resource discounts, free trainings, and collaborative marketing opportunities

Hot
Try Now

GaussDB(DWS) FAQs

  • Why Should I Use Public Cloud GaussDB(DWS)?

    Conventional data warehouses are not practical for smaller enterprises due to high cost, time-consuming device and system selection and procurement, and complex scale-out.

    GaussDB(DWS) on the public cloud is the better choice:

    1. This cloud service of distributed MPP data warehousing is very open, efficient, compatible, scalable, and is easy to O&M.
    2. Developed on the FusionInsight LibrA data warehouse kernel, it empowers public cloud enterprises with better and consistent experience on and off the cloud.
    3. FusionInsight LibrA is a next-generation distributed data warehousing system with independent intellectual property rights. Currently, it is widely used in government, finance, and carriers. FusionInsight LibrA is compatible with mainstream open-source Postgres databases, especially in Oracle and Teradata SQL statements. Our data warehouse engineers have designed a kernel of hybrid row-column stores not only for faster analysis but also for data processing, such as adding, deleting, and modifying data. FusionInsight LibrA features the cost optimizer and warehouse technologies, including machine code vector computing and inter/intra-parallelism for operators and nodes. It uses LLVM to optimize the local code in compilation query plans. More powerful data query and analysis addresses service pain points and improves user experience.
    4. Application is out of the box and takes just a few minutes so you no longer need to look for a warehouse to buy. Simpler procurement and lower cost and barriers give smaller enterprises access to more value from their data for enterprise development and insights.
  • When Should I Use GaussDB(DWS) and MRS?

    MRS works better with big data processing frameworks such as Apache Spark, Hadoop, and HBase, to process and analyze ultra-large data sets through custom code. It allows you to control cluster configurations and software installed in the cluster.

    GaussDB(DWS) works better with complex queries of a large amount of structured data. It aims to pool data from different sources together, such as inventory, finance, and retail system. To ensure consistency and accuracy of enterprise reports, GaussDB(DWS) stores data in a highly structured manner. This structure can directly build the data consistency rule to the database table. Additionally, GaussDB(DWS) is highly compatible with standard SQL statements and the syntax of conventional transaction-supported databases.

    GaussDB(DWS) is preferred when you want to perform complex query of a large amount of structured data with high performance.

  • What Are the Differences Between GaussDB(DWS) and Hive in Functions?

    GaussDB(DWS) and Hive have different functions in the following aspects:

    1. Hive is a data warehouse based on Hadoop MapReduce. GaussDB(DWS) is a data warehouse based on Postgres MPP.
    2. Hive data is stored on HDFS. GaussDB(DWS) data can be stored locally or on OBS in foreign table form.
    3. Hive does not support indexes. GaussDB(DWS) supports indexes, so querying is faster.
    4. Hive does not support stored procedures. GaussDB(DWS) does, so it has more extensive application scenarios.
    5. Hive supports fewer SQL statements than GaussDB(DWS), including functions, customized functions, and stored procedures.
    6. Hive does not support transactions. GaussDB(DWS) supports complete transactions.
    7. Both Hive and GaussDB(DWS) support backups, so the reliability is the same.
    8. GaussDB(DWS) delivers much better performance than Hive.

    Based on their respective functions, Hive is useful for offline analysis while GaussDB(DWS) is useful for both online analysis and ad-hoc query.

  • How Is GaussDB(DWS) Secured?

    GaussDB(DWS) uses IAM and VPC to control user access and isolate cluster network. Cluster access is over SSL and cipher suite. Additionally, GaussDB(DWS) supports two-way digital certificate authentication.

    Node OSs in each cluster are hardened to allow valid access to only OS files.

  • Can I Modify the Security Group of a GaussDB(DWS) Cluster?

    Yes. When a data warehouse cluster is created, its security group cannot be changed. However, you can add, delete, or modify rules of the current security group.

    To edit the cluster security group:

    1. Log in to the GaussDB(DWS) console.
    2. In the navigation pane on the left, choose Clusters.
    3. In the cluster list, find the target cluster and click the cluster name. The Basic Information page is displayed.
    4. Locate the Security Group parameter and click the security group name to switch to the Security Groups page on the VPC console, on which you can set the security group.
  • How Are LibrA, GaussDB A, and GaussDB(DWS) Related?

    GaussDB(DWS) is an online data processing database built on the public cloud infrastructure and platform. It evolved from Huawei's own GaussDB A (originally called FusionInsight LibrA). GaussDB A is a database software deployed on physical machines. For more information, visit the following websites:

    1. Version 6.5.1 or earlier: https://support.huawei.com/enterprise/en/cloud-computing/gaussdb-200-pid-21407429
    2. Version 8.0.0: https://support.huawei.com/enterprise/en/cloud-computing/gaussdb-a-pid-250949677
  • What Do I Do If Creating a GaussDB(DWS) Cluster Failed?

    Troubleshooting

    Check that you have enough quota for creating the cluster.

    Technical Support

    If the fault cannot be identified, submit a service ticket to report the problem: Log in to the console and choose Service Tickets > Create Service Ticket.

  • Does GaussDB(DWS) Support a Single Node for a Learning Environment?

    Yes. In GaussDB(DWS), you can create a hybrid data warehouse cluster in standalone mode. If the name of the selected node flavor contains h1 (for example, dwsx2.h1.xlarge.2.c6), the hybrid data warehouse only supports standalone deployment, which does not provide HA capabilities. The storage cost can be reduced by half. A standalone data warehouse can be restored by the automatic reconstruction of ECS, and its data reliability is ensured by the EVS multi-copy mechanism. It is less expensive than other specifications. It is a good choice for lightweight services.

  • Does GaussDB(DWS) Support BMS?

    Yes. You can submit a service ticket to apply for BMS flavors. GaussDB(DWS) uses ECSs only on Huawei Cloud by default.

  • How Is the Disk Space or Capacity of GaussDB(DWS) Calculated?

    A cluster with three data nodes of 320 GB each has a total capacity of 960 GB. When 1 GB data is stored, GaussDB(DWS) stores 1 GB data on two nodes due to duplication, a security mechanism, thereby occupying a total of 2 GB space. As a result, more than 2 GB space is occupied if metadata and indexes are calculated. Therefore, a three-node cluster with a total capacity of 960 GB can store 480 GB data. This mechanism ensures data security.

    When you purchase nodes on the console, you are billed by the available capacity of a node. For example, the actual space of dws.m3.xlarge is 320 GB and the available space displayed is 160 GB, the space you will be billed for.

We use cookies to improve our site and your experience. By continuing to browse our site you accept our cookie policy. Find out more