This is an automated email from the ASF dual-hosted git repository.
victoria pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/druid.git
The following commit(s) were added to refs/heads/master by this push:
new a2be685824 Remove the time bit, fix headings (#12808)
a2be685824 is described below
commit a2be6858248224ab103dc31e7576a63e6b143e40
Author: Katya Macedo <[email protected]>
AuthorDate: Wed Jul 20 17:37:57 2022 -0500
Remove the time bit, fix headings (#12808)
* Remove the time bit, fix headings
* Adopt review suggestions
* Edits
* Update smoosh file description
* Adopt review suggestions
* Update spelling
---
docs/design/segments.md | 15 +++++++--------
website/.spelling | 2 ++
2 files changed, 9 insertions(+), 8 deletions(-)
diff --git a/docs/design/segments.md b/docs/design/segments.md
index 6e6a5615a9..048cf39696 100644
--- a/docs/design/segments.md
+++ b/docs/design/segments.md
@@ -28,7 +28,7 @@ Apache Druid stores its data and indexes in *segment files*
partitioned by time.
The time interval is configurable in the `segmentGranularity` parameter of the
[`granularitySpec`](../ingestion/ingestion-spec.md#granularityspec).
For Druid to operate well under heavy query load, it is important for the
segment
-file size to be within the recommended range of 300MB-700MB. If your
+file size to be within the recommended range of 300-700 MB. If your
segment files are larger than this range, then consider either
changing the granularity of the segment time interval or partitioning your
data and/or adjusting the `targetRowsPerSegment` in your `partitionsSpec`.
@@ -40,7 +40,7 @@ for more guidance.
## Segment file structure
Segment files are *columnar*: the data for each column is laid out in
-separate data structures. By storing each column separately, Druid decreases
query latency by scanning only those columns actually needed for a query.
There are three basic column types: timestamp, dimensions, and metrics:
+separate data structures. By storing each column separately, Druid decreases
query latency by scanning only those columns actually needed for a query. There
are three basic column types: timestamp, dimensions, and metrics:
@@ -189,15 +189,14 @@ A segment contains several files:
* `meta.smoosh`
- A file containing metadata (filenames and offsets) about the contents of
the other `smoosh` files
+ A file containing metadata (filenames and offsets) about the contents of
the other `smoosh` files.
* `XXXXX.smoosh`
- A number of files containing concatenated binary data.
-
- The `smoosh` files represent multiple files "smooshed" together in order
to minimize the number of file descriptors that must be open to house the data.
They are files of up to 2GB in size (to match the limit of a memory mapped
ByteBuffer in Java). The `smoosh` files house individual files for each of the
columns in the data as well as an `index.drd` file with extra metadata about
the segment.
-
-Additionally, a column called `__time` refers to the time column of the
segment.
+ Smoosh (`.smoosh`) files contain concatenated binary data. This file
consolidation reduces the number of file descriptors that must be open when
accessing data. The files are 2 GB or less in size to remain within the limit
of a memory-mapped `ByteBuffer` in Java.
+ Smoosh files contain the following:
+ - Individual files for each column in the data, including one for the
`__time` column that refers to the timestamp of the segment.
+ - An `index.drd` file that contains additional segment metadata.
In the codebase, segments have an internal format version. The current segment
format version is `v9`.
diff --git a/website/.spelling b/website/.spelling
index 84a523e141..3db7cb020e 100644
--- a/website/.spelling
+++ b/website/.spelling
@@ -413,6 +413,8 @@ simple-client-sslcontext
sharded
sharding
skipHeaderRows
+Smoosh
+smoosh
smooshed
splittable
ssl
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
