Elastic has announced version 7.12 of its Elasticsearch search engine. The update brings a number of improvements, including support for runtime fields, a new frozen tier for storage and support for the ARM architecture.
An overview of the new features can be found in a blog post from Elastic. The company opens its post with information on runtime fields. This is a new technique where fields do not need to be defined in advance. This should offer more flexibility in storing data. However, this flexibility comes at the cost of speed, so the developer must weigh up whether it is a better technique than the classic indexed fields.
For the cheap storage of large amounts of data, Elastic has announced the frozen tier. Here, the storage is not directly connected to compute but stored in bulk in object stores such as Amazon S3, Microsoft Azure Storage and Google Cloud Storage. Such bulk storage should be much cheaper than local storage, with a compromise in search speed. Data that needs to be searchable faster can still be stored locally.
To assist developers in moving their data to cloud storage, Elastic has made a repository test kit available. With it, developers can validate whether their S3-compatible object store is suitable for searchable snapshots, the cold tier and the frozen tier.
Furthermore, Elastic has now announced official support for the Arm aarch64 architecture for both Elasticsearch and Kibana. This is in response to the general trend towards Arm processors in data centres. The company promises that all software should be able to run entirely on Arm and promises the same support for developers as is provided for x86, both by the support team and the community forum. In the future, the company also plans to make Elastic Cloud work on Arm.
Finally, Elastic has announced some smaller changes. For example, the memory usage of the search engine should be improved, administrators will be able to remove old async searches and changes have been made to geoshapes.
Elasticsearch 7.12 is available now.