Create data box

Prev Next

Available in VPC

Once you have completed Cloud Data Box subscription, you can create a data box. During the process, you have to configure analysis services for your data box and servers to access the analysis services.

1. Select infrastructure product

To use a data box, you have to select analysis services such as Cloud Hadoop or TensorFlow and Connect Server to access the analysis services.

  • Connect Server: a server supporting Windows OS. Using PuTTY on the Connect Server, you can access Cloud Hadoop or Ncloud TensorFlow Server. You can also access Ambari, Zeppelin or Jupyter Notebook through a web browser on Connect Server.
  • SSL VPN: to access the offered servers, you must use SSL VPN.
  • NAS: used to import files from external sources or to export analysis results after external communications are blocked. The NAS comes mounted on Cloud Hadoop, TensorFlow CPU, TensorFlow GPU, and Linux server, so you can easily share files between Cloud Hadoop, Ncloud TensorFlow Server, and Linux server.

To create a data box, follow these steps:

  1. Click i_menu > Services > Big Data & Analytics > Cloud Data Box on the NAVER Cloud Platform console.
  2. Click the [Request data box] button.
    • A screen where you can configure the infrastructure environment for data analysis appears.
  3. Select the number and the specifications of Connect Server.
    • Each Connect Server can have 2 user accounts can be accessed by up to 2 users simultaneously. Decide how many servers you will need based on the number of your analysts and select a number accordingly.
  4. Select the number of SSL VPN accounts you need.
    • As email or mobile authentication is mandatory when accessing SSL VPN, make sure to apply for 1 SSL VPN account per user.
    • You can apply for up to 3 times the number of the Connect Servers.
  5. Enter capacity and quantity of the NAS you want to create and click the [Add] button.
    • The base volume capacity for NAS ranges from 500 to 10,000 GB, and you can increase it by 100 GB.
    • If you are reusing the established NAS volume, you do not have to create a new one. If you are not reusing the established NAS, you have to create more than 1 NAS.
    • You can create up to 4 NASs.
  6. Go to the Cloud Hadoop area and select edge nodes and master nodes.
    • The number of edge nodes and master nodes is fixed and cannot be altered.
  7. Go to the Cloud Hadoop area, select worker nodes, and set its quantity.
    • Once the data box creation process has been completed, the quantity of worker nodes cannot be altered.
  8. Select the number and the specifications of Ncloud TensorFlow Server.
    • You must apply for at least 1 Ncloud TensorFlow Server (CPU) or 1 Ncloud TensorFlow Server (GPU).
  9. If you want to create an additional Linux server, select its OS, specifications, and quantity and click the [Add] button.
  10. Check the application details and click the [Next] button.
    clouddatabox-add_new_1_ko
Note
  • If you had returned a data box within the last 7 days, Reuse previous data box's NAS feature is enabled. Click the checkbox of Reuse previous data box's NAS to reuse the NAS volume of the previous data box.
  • The previous data box's NAS is not made available the moment the data box creation is completed. You can only access it after the communication with external networks is blocked following the data supply request submission.

2. Enter product information

This is the stage where you enter the name of the data box and configure the access details of the Connect Server, the Cloud Hadoop, and the TensorFlow (CPU/GPU) server. As account name for each server is generated automatically, you only need to create their passwords. To name the data box and configure access details, follow these steps:

  1. Enter the data box's name.
    • Only Korean, Roman alphabet, numbers, and hyphen (-) are allowed. Enter 3 to 20 letters.
  2. Enter the Connect Server's password.
    • 2 accounts are created per Connect Server.
    • The accounts are named "ncp1" and "ncp2".
  3. Enter Hadoop cluster's password.
    • Cloud Hadoop's account is named "ncp".
  4. Enter TensorFlow (CPU/GPU) Server's password.
    • TensorFlow's account is named "root".
  5. Enter the added Lunix server's password.
    • Linux server's account is named "root".
  6. Check the input details and click the [Next] button.
    clouddatabox-add_new_2_ko

3. Final confirmation

This is the stage where you review and confirm the application details and submit the request for data box creation. It may take several hours or more for the data box to be created. Once the data box is created, you will be notified via email. To review the data box creation application details, follow these steps:

  1. Check if the data box's name has been properly spelled and registered for application.
  2. Check if the infrastructure services have been applied for.
  3. Review the details and click the [OK] button.