What’s New in Couchbase

Key new features

Couchbase Enterprise Server 7.2 supports time-series data in JSON

Time-series data is both highly structured and voluminous – this type of data is often able to overwhelm traditional databases with rows and rows of measurements taken at frequent intervals. Couchbase has taken a novel approach to supporting time-series data by ingesting and fitting it into JSON array structures. This allows the entire Couchbase product line the ability to offer time-series data access without increasing the complexity of the database’s operation or behavior.

Time-series support also adds new functions for pre-formatting the data prior to ingestion into a time-series array, as well as functions for extracting and parsing that data for use in applications or analytics.

There are additional benefits of using JSON as the time-series container in Couchbase, including fast processing and retrieval, use of a single index to access the array, high-density storage with improved compaction, and flexible assembly and decoding of the array for use in applications and analytics.

An example of JSON time series data for a fictional stock ticker
{
"ticker": "XYZ",
"ts_interval": 86400000,
"ts_start": 1359676800000,
"ts_end": 1362009600000,
"ts_data": [
[ 27.285, 27.595, 27.24, 27.295 ],
[ 27.64, 27.95, 27.365, 27.61 ],
// ...
[ 27.45, 27.605, 27.395, 27.545 ]
]
}


The introduction of time-series support in Couchbase has also driven the expansion of development in the underlying storage operation of Couchbase Magma, the high-density key-value store that underpins the Couchbase database. For this release, we have introduced new block size configuration settings to make disk compression more efficient. Block sizes by default are set to 4096 bytes and are now adjustable to a maximum of 131072 bytes. Using larger block sizes may result in greater compression. An estimation utility is also available in Magma to advise users on block size compression ratios to expect, given their document capacity expectations.

Enterprise Server 7.2 adds change data capture (CDC)

Server 7.2 offers the ability to log and stream all document changes in a collection or bucket via Kafka. This allows developers to use a document’s change history as programmatic inputs to their application. Use cases for this feature include:

• Developing a persistent audit trail for data changes
• Triggering downstream business processes that listen for these changes via Kafka
• Expanding future development with integrations to include feeding the Eventing services with change history as well as creating a change stream API for Couchbase SDKs

Enterprise Server 7.2 adds cost-based optimization (CBO) for Analytics

Couchbase Analytics service has adopted the same CBO features available in the SQL++ Query Service. This patented optimization utility will help analytic queries run as efficiently as possible. This optimizer uses a sample-based approach to support complex analytic queries. Sampling is drawn from the Analytics collection metadata during the query planning stage of query execution.

Enterprise Server 7.2 adds Reliability, Availability, and Serviceability improvements

• Alert for Low Indexer Resident Percentage when Index service resources are nearing critical levels.
• Index Redistribution flag for partitioned indexes. This flag allows finer control over when indexes and partition indexes are moved during rebalances.
• Expanded status metrics for Data Service warmup via the Prometheus open metric endpoint to feed alerting dashboards.
• Expanded metrics for inbound replication activity in cross data center replication (XDCR).
• TLS certificate management improvements including automatic validation checking for the Subject Alternative Names (SAN) hosts information, inclusion of the Capella CA certificate in Server 7.2 clusters to facilitate hybrid XDCR support out of the box, and the depreciation of support for TLS 1.0 and 1.1.
• Support for current Linux operating systems is added for Red Hat RHEL 9, Oracle Linux 9, AWS Linux 2023, Ubuntu 22 LTS, and Apple MacOS 12.
• Support for the following operating systems has been dropped: RHEL 7, CentOS 7, Oracle Linux 7, Ubuntu 18 LTS, SUSE Linux 12 versions older than SP2, and macOS 10.15.

Scopes and collections in Couchbase Mobile enables you to:

• With high-density storage, data capacity per node is extended from 3TB to 10TBs
• Performance tests exhibit 4x higher throughput (operations per second) and improvements to both write and read speeds
• High-density storage is also designed to support and scale with high-end SSDs such as non-volatile memory express (NVMe) SSDs
• Hardware specifications lower the per-node memory-to-data ratio from 10% to 1%. This allows customers to scale up to 10x more using their existing infrastructure, or to reduce by up to a factor of 10 the number of cluster nodes they need to handle large volumes of data.

The ability to manage and enforce independent app lifecycles, as well as leverage fine-grained security access control, will dramatically simplify the deployment of multi-tenant and high-application density cloud-to-edge applications at scale.

By providing better and more granular organization of data, security, and synchronization, scopes and collections in Couchbase Mobile 3.1 simplifies apps and makes them more secure, scalable, and efficient.

Key capabilities

Multi-tenant mobile applications

Consolidate multiple tenant buckets to scopes within a single bucket. Reduce cluster footprint by orders of magnitude.

High density cloud-to-edge apps

Scope-based microservices can serve multiple applications from a single bucket. Each application is associated with its own scope.

Simplified data organization

Share user management across scopes, simplify access control to “per collection”, and provide easier migration from relational systems where schemas and tables map directly to scopes and collections.

Fine-grained end-to-end data sync

Augment channel-based access control and filtering support with the ability to sync data at the granularity of scopes or collections.

Embedded application data isolation

Isolate on-device app data into scopes on a single embedded database versus multiple embedded databases for local only or sync data.

Sync metadata isolation

Isolate gateway system-level metadata from application data, simplifying queries and making app development more manageable and straightforward.