Predix Search Overview

About Predix Search

Predix Search adds powerful search capabilities to your applications.

Predix Search stores and searches big data in near real time. Powered by Elasticsearch (version 6.2), Predix Search exposes an extensive REST API accessible through a variety of programming languages, including Java, Python, PHP, .NET, and SQL.

The API enables you to:
  • Index documents formatted in JSON
  • Search the contents of indices
  • Perform advanced search operations such as sorting, filtering, scripting, pagination, and aggregation
  • Update and delete indices
  • Check cluster, node, and index status and statistics

Predix Search Clusters

Predix Search clusters deliver high performance by enabling federated indexing and searching across the individual nodes that make up the server cluster.

Dedicated Master Clusters

Predix Search includes the option of adding a dedicated master cluster to increase overall system stability. Dedicated master clusters offload cluster management tasks to master nodes. Because master nodes do not take part in indexing and search-related data processing, system stability is enhanced by creating a separation of responsibilities between the master nodes and data nodes in the system. At least one dedicated master cluster is mandatory if the total number of nodes is more than 10.

Dedicated master nodes manage clusters by:
  • Tracking all nodes in the cluster
  • Tracking the number of indices in the cluster
  • Tracking the number of Elasticsearch shards belonging to each index
  • Maintaining routing information for nodes in the cluster
  • Updating the cluster state after state changes, such as the creation of an index or addition or deletion of nodes in the cluster
  • Replicating changes to the cluster state across all nodes in the cluster
  • Monitoring the health of nodes by sending¬†heartbeat signals, periodic signals that monitor the availability of the data nodes in the cluster

Multiple Availability Zones

Predix Search ensures data availability and durability by enabling cluster deployment across availability zones (AZs). AZs are physically separate computing infrastructures designed for reliability. In the event of an infrastructure failure in the primary AZ, Predix Search automatically fails over to the standby AZ.

Multiple availability zone support (Multi-AZ) is an optional feature of Predix Search. Multi-AZ requires an even number of data nodes; half of the data nodes in the primary AZ, half in the standby. In Multi-AZ implementations, Elasticsearch shards are stored on nodes in the primary AZ, replicas on nodes in the standby.

Additional Features

  • Data nodes scale to a maximum of 10 for the Dedicated-E10 plan, 20 for the Dedicated-E20 to Dedicated-E80 plans.
  • Daily backups are archived for up to 14 days.