Step 1: Add cluster details - Amazon FinSpace

Step 1: Add cluster details

Specify details for each of the following sections on Add cluster details page.

Cluster details

  1. Choose from one of the following types of clusters that you want to add.

    • (HDB) Historical Database

    • (RDB) Realtime Database

    • Gateway

    • General purpose

    • Tickerplant

    For more information about cluster types, see Managed kdb Insights clusters.

    Note
    • Currently, you can only create dedicated HDB clusters and all cluster types that are running on a scaling group directly from the console. To create other types of clusters, you need to first create a Support case, and then proceed with steps in this tutorial.

    • The parameters that are displayed on the Step 5: Configure data and storage page will change based on the cluster type and running mode that you select in this step.

  2. Add a unique name and a brief description for your cluster.

  3. For Release label, choose the package version to run in the cluster.

  4. (Optional) Choose the IAM role that defines a set of permissions associated with this cluster. This is an execution role that will be associated with the cluster. You can use this role to control access to other clusters in your Managed kdb environment.

Cluster running mode

  1. Choose if you want to add this cluster as a dedicated cluster or as a part of scaling groups.

    • Run on kdb scaling group – Allows you to share a single set of compute with multiple clusters.

    • Run as a dedicated cluster – Allows you to run each process on its own compute hose.

  2. If you choose Run as a dedicated cluster, you also need to provide the Availability Zones where you want to create a cluster.

    1. Choose AZ mode to specify the number of Availability Zones where you want to create a cluster. You can choose from one of the following options:

      • Single – Allows you to create a cluster in one Availability Zone that you select. If you choose this option, you must specify only one Availability Zone value and only one subnet in the next step. The subnet must reside in one of the three AZs that your kdb environment uses, and the Availability Zone must align with one of the three AZs.

      • Multiple – Allows you to create a cluster with nodes automatically allocated across all the Availability Zones that are used by your Managed kdb environment. This option provides resiliency for node or cache failures in a Single-AZ. If you choose this option, you must specify three subnets, one in each of the three AZs that your kdb environment uses.

      Note

      For the General purpose and Tickerplant type cluster, you can only choose Single-AZ.

    2. Choose the Availability Zone IDs that include the subnets you want to add.

Scaling group details

Note

This section is only available when you choose to add cluster as a part of scaling groups.

Choose the name of the scaling group where you want to create this cluster. The drop down shows the metadata for each scaling group along with their names to help you decide which one to pick. If a scaling group is not available, choose Create kdb scaling group to add a new one. For more information, see Creating a Managed kdb scaling group.

Node details

In this section, you can choose the capacity configuration for your clusters. The fields in this section vary for dedicated and scaling group clusters.

Scaling group cluster

You can decide the memory and CPU usage that will be shared with the instances for your scaling group clusters by providing the following information.

  1. Under Node details, for Node count, enter the number of instances in a cluster.

    Note

    For a General purpose and Tickerplant type cluster, the node count is fixed at 1.

  2. Enter the memory reservation and limits per node. Specifying the memory limit is optional. The memory limit should be equal to or greater than the memory reservation.

  3. (Optional) Enter the number of vCPUs that you want to reserve for each node of this scaling group cluster.

Dedicated cluster

For a dedicated cluster you can provide an initial node count and choose the capacity configuration from a pre-defined list of node types. For example, the node type kx.s.large allows you to use two vCPUs and 12 GiB of memory for your instance.

  1. Under Node details, for Initial node count, enter the number of instances in a cluster.

    Note

    For a General purpose and Tickerplant type cluster, the node count is fixed at 1.

  2. For Node type, choose the memory and storage capabilities for your cluster instance. You can choose from one of the following options:

    • kx.s.large – The node type with a configuration of 12 GiB memory and 2 vCPUs.

    • kx.s.xlarge – The node type with a configuration of 27 GiB memory and 4 vCPUs.

    • kx.s.2xlarge – The node type with a configuration of 54 GiB memory and 8 vCPUs.

    • kx.s.4xlarge – The node type with a configuration of 108 GiB memory and 16 vCPUs.

    • kx.s.8xlarge – The node type with a configuration of 216 GiB memory and 32 vCPUs.

    • kx.s.16xlarge – The node type with a configuration of 432 GiB memory and 64 vCPUs.

    • kx.s.32xlarge – The node type with a configuration of 864 GiB memory and 128 vCPUs.

Auto-scaling

Note

This section is only available when you add an HDB cluster type as a dedicated cluster.

Specify details to scale in or scale out the based on service utilization. For more information, see Auto scaling.

  1. Enter a minimum node count. Valid numbers: 1–5.

  2. Enter a maximum node count. Valid numbers: 1–5.

  3. Choose the metrics to auto scale your cluster. Currently, FinSpace only supports CPU utilization.

  4. Enter the cooldown time before initiating another scaling event.

Tags

  1. (Optional) Add a new tag to assign to your kdb cluster. For more information, see AWS tags.

    Note

    You can only add up to 50 tags to your cluster.

  2. Choose Next for next step of the wizard.