Features

Rapid Deployment

CSS enables you to create Elasticsearch, OpenSearch, or Logstash clusters ready to run enterprise workloads in a few simple steps. CSS will automatically install and deploy a search engine based on the cluster type, version, and node specifications you select.

• Elasticsearch clusters are suited for log analytics, enterprise search, and real-time data visualization, particularly where a robust analytics ecosystem and strong community support are required. For more information, see Creating an Elasticsearch Cluster.
• OpenSearch clusters are an ideal choice for customers who require elastic scalability, continuous feature iteration, and long-term technology evolution. For more information, see Creating an OpenSearch Cluster.
• Logstash supports a wide range of use cases, including log management, data integration, and real-time data processing. For more information, see Creating a Logstash Cluster.

Enhanced Kernel

CSS features an enhanced kernel that boosts cluster performance and stability, while optimizing costs across diverse deployment scenarios.

• Vector search: Unstructured data, such as images, videos, and language corpora, is converted into vectors, which are searched based on similarity using either an exact or approximate nearest neighbors algorithm.
• Decoupled storage and compute: CSS stores new data (hot data) on SSDs to ensure optimal query performance, and historical data (cold data) in OBS to cut storage costs.
• Flow control: Flow control policies, such as client request throttling, shard indexing backpressure, and traffic pattern analysis clusters, protect clusters from overload, ensuring proper resource allocation and risk prevention.
• Large query isolation: Queries that have high memory usage or take too long to complete are suspended or isolated to improve cluster stability and prevent out-of-memory (OOM) exceptions.
• Enhanced aggregation: CSS enhances aggregation performance in the face of large data volumes by leveraging vectorization and optimized clustering techniques, enabling faster analytics and decision-making in complex situations.
• Read/write splitting: This feature relies on collaboration between a pair of leader and follower clusters, where the leader exclusively handles write requests, and the follower serves read queries. This separation of workloads enhances the stability, scalability, and high-concurrency processing capabilities of clusters.
• Switching between hot and cold storage: Data is allocated to nodes of different performance standards based on data temperature (that is, how often data is accessed). The goal is to achieve optimal storage costs and query performance. Typically, high-performance hardware (such as SSDs) is used to store hot data that is frequently accessed, while inexpensive hardware (such as HDDs) is used to store cold data that is infrequently accessed.
• Index recycle bin: To prevent data loss caused by accidental deletion, CSS provides an index recycle bin. When it is enabled, deleted indexes are temporarily stored in the recycle bin, allowing for recovery before they are permanently removed. This feature improves data reliability and operational security.
• Enhanced data ingestion: This feature significantly enhances data ingestion performance and reduces write rejections through bulk route optimization, text index acceleration, and word segmentation acceleration.
• Index monitoring: This feature helps you monitor the index status, statistics, and trends of clusters, allowing you to promptly handle potential risks to ensure cluster reliability.

For more information, see Search Enhancement Features for Elasticsearch Clusters or Search Enhancement Features for OpenSearch Clusters.

Public Network Access

In CSS, you can enable public network access for Elasticsearch and OpenSearch clusters that have the security mode enabled and use HTTPS. With public network access enabled, a cluster is automatically assigned a public IP address with dedicated dynamic BGP bandwidth, making it accessible from the Internet via HTTPS.

For more information, see Configuring Public Network Access for an Elasticsearch Cluster or Configuring Public Network Access for an OpenSearch Cluster.

VPC Endpoint

VPC Endpoint enables you to access resources across Virtual Private Clouds (VPCs) using a dedicated gateway, without exposing the network information of servers. When VPC Endpoint is enabled for an Elasticsearch or OpenSearch cluster, a VPC endpoint is automatically allocated to it, through which users will be able to access this cluster.

For more information, see Configuring VPC Endpoints for an Elasticsearch Cluster or Configuring VPC Endpoints for an OpenSearch Cluster.

Dedicated Load Balancers

In CSS, you can configure dedicated load balancers for clusters. Compared with shared load balancers, dedicated load balancers provide higher performance and more diverse features. If your Elasticsearch or OpenSearch cluster contains a large number of nodes and stores large quantities of data, you are advised to configure load balancers for it to improve stability.

For more information, see Configuring a Dedicated Load Balancer for an Elasticsearch Cluster or Configuring a Dedicated Load Balancer for an OpenSearch Cluster.

Backup and Restoration

In CSS, you can use snapshots to back up and restore the data of Elasticsearch and OpenSearch clusters. By storing a snapshot in an OBS bucket, you save a point-in-time copy of the cluster's data. Later, you can restore this snapshot to the original cluster or another eligible cluster.

This process fetches the backed-up index data, metadata, and shard structure from OBS and reconstruct them in the selected destination cluster, ensuring data consistency. This capability is essential for operations like data rollback, cluster migration, and disaster recovery.

For more information, see Backing up and Restoring the Data of an Elasticsearch Cluster or Backing up and Restoring the Data of an OpenSearch Cluster.

Flexible Scalability

CSS clusters support flexible scaling options, allowing you to dynamically adjust capacities to meet changing demand, ensuring optimal performance and efficiency.

• Scaling out or up: If a cluster struggles to maintain performance in the face of rapid data growth or sustained high memory usage, you may scale it horizontally by adding more nodes and more node types, or do so vertically by increasing the storage capacity of existing nodes.
• Scaling in: If a cluster has excess capacity due to off-peak traffic or reduced data volume, you can reduce its nodes to optimize costs.
• Node specification changes: If the workloads on a cluster's data plane change or the cluster is experiencing performance bottlenecks, you can scale the cluster vertically by changing its node specifications or node storage type.
• Replacing a specified node: You can use this option to replace failed cluster nodes.
• Changing the security mode: You can enable or disable the security mode for an existing cluster to accommodate changing security needs.
• Switching AZs: CSS supports cross-AZ deployment. By switching AZs for a cluster, you can make more resources available to the cluster or improve the cluster's availability.

Upgrade

In CSS, you can upgrade clusters to use feature and performance enhancements or fix known issues.

• Same-version upgrade: Install kernel patches for a cluster. The cluster is upgraded to the latest image of the current version to fix known issues or optimize performance.
• Cross-version upgrade: Upgrade the cluster version. The cluster will be upgraded to the latest image of the target version to enhance performance or incorporate older versions.

For more information, see Upgrading an Elasticsearch Cluster or Upgrading an OpenSearch Cluster.

Log Management

CSS provides log ingestion, log query, and log backup, enabling you to easily manage and analyze logs to efficiently locate faults, optimize performance, and enhance system security.

• Log query: On the log management page of the CSS management console, you can query the latest log records by node, log type, and other criteria, so you can quickly locate or diagnose issues.
• Log backup: Cluster logs are periodically synchronized to OBS buckets. You can download them for in-depth analysis at any time. Custom backup policies (such as backup time and storage path) are supported.
• Log ingestion (for Elasticsearch and OpenSearch clusters only): Cluster logs are transmitted in real time to the current cluster or another cluster on the same network. (Certain version and network compatibility requirements must be met.) You can use Kibana, OpenSearch Dashboards, or other visualization tools for log search and analytics. You can also set index prefixes and retention periods to flexibly manage the log lifecycle. Logs can be transferred across clusters, facilitating centralized monitoring in a multi-cluster environment.

For more information, see Querying and Managing Elasticsearch Cluster Logs or Managing OpenSearch Cluster Logs.

Monitoring and Alerting

Through Cloud Eye, you can monitor the health status of CSS clusters by checking alarms and events generated for them. You can customize alarm thresholds for individual metrics. When such thresholds are reached, alarms will be triggered automatically. CSS can also interconnect with the Simple Message Notification (SMN) service to push alarm notifications to users via text messages or email.