data-streams.asciidoc 2.1 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
  1. [[data-streams]]
  2. = Data streams
  3. ++++
  4. <titleabbrev>Data streams</titleabbrev>
  5. ++++
  6. A _data stream_ is a convenient, scalable way to ingest, search, and manage
  7. continuously generated time-series data.
  8. Time-series data, such as logs, tends to grow over time. While storing an entire
  9. time series in a single {es} index is simpler, it is often more efficient and
  10. cost-effective to store large volumes of data across multiple, time-based
  11. indices. Multiple indices let you move indices containing older, less frequently
  12. queried data to less expensive hardware and delete indices when they're no
  13. longer needed, reducing overhead and storage costs.
  14. A data stream is designed to give you the best of both worlds:
  15. * The simplicity of a single, named resource you can use for requests
  16. related
  17. * The storage, scalability, and cost-saving benefits of multiple indices
  18. You can submit indexing and search requests directly to a data stream. The
  19. stream automatically routes the requests to a collection of hidden,
  20. auto-generated indices that store the stream's data.
  21. You can use a <<indices-templates,composable template>> and
  22. <<index-lifecycle-management,{ilm} ({ilm-init})>> to automate the management of
  23. these hidden indices. You can use {ilm-init} to spin up new indices, allocate
  24. indices to different hardware, delete old indices, and take other automatic
  25. actions based on age or size criteria you set. This lets you seamlessly scale
  26. your data storage based on your budget, performance, resiliency, and retention
  27. needs.
  28. [discrete]
  29. [[when-to-use-data-streams]]
  30. == When to use data streams
  31. We recommend using data streams if you:
  32. * Use {es} to ingest, search, and manage large volumes of time-series data
  33. * Want to scale and reduce costs by using {ilm-init} to automate the management
  34. of your indices
  35. * Index large volumes of time-series data in {es} but rarely delete or update
  36. individual documents
  37. [discrete]
  38. [[data-streams-toc]]
  39. == In this section
  40. * <<data-streams-overview>>
  41. * <<set-up-a-data-stream>>
  42. * <<use-a-data-stream>>
  43. include::data-streams-overview.asciidoc[]
  44. include::set-up-a-data-stream.asciidoc[]
  45. include::use-a-data-stream.asciidoc[]