From e022a727a8f165c2f9f04443be5f43e659b572d3 Mon Sep 17 00:00:00 2001 From: Raul Leite Date: Fri, 14 Nov 2025 15:31:21 -0600 Subject: [PATCH 1/3] =?UTF-8?q?I=E2=80=99ve=20proposed=20a=20slight=20rewo?= =?UTF-8?q?rding=20of=20this=20section=20to=20improve=20clarity=20and=20re?= =?UTF-8?q?adability.=20(On-Disk=20Layout=20Paragraph)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Signed-off-by: Raul Leite --- docs/storage.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/docs/storage.md b/docs/storage.md index 7b6e3bffe8..de8a8ed065 100644 --- a/docs/storage.md +++ b/docs/storage.md @@ -11,13 +11,13 @@ Prometheus's local time series database stores data in a custom, highly efficien ### On-disk layout -Ingested samples are grouped into blocks of two hours. Each two-hour block consists -of a directory containing a chunks subdirectory containing all the time series samples -for that window of time, a metadata file, and an index file (which indexes metric names -and labels to time series in the chunks directory). The samples in the chunks directory -are grouped together into one or more segment files of up to 512MB each by default. When -series are deleted via the API, deletion records are stored in separate tombstone files -(instead of deleting the data immediately from the chunk segments). +Ingested samples are grouped into two-hour blocks. Each block consists of a directory that +contains a chunks subdirectory with all the time series samples for that time window, +a metadata file, and an index file (which maps metric names and labels to the time series +in the chunks directory). By default, the samples in the chunks directory are organized +into one or more segment files, each up to 512 MB. When series are deleted via the API, +deletion records are stored in separate tombstone files rather than being immediately +removed from the chunk segments. The current block for incoming samples is kept in memory and is not fully persisted. It is secured against crashes by a write-ahead log (WAL) that can be From c9827ef983038183a99b5ee851dc50332b29392e Mon Sep 17 00:00:00 2001 From: Raul Leite Date: Sat, 15 Nov 2025 10:52:18 -0600 Subject: [PATCH 2/3] Fix formatting in storage.md extra space removed Signed-off-by: Raul Leite --- docs/storage.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/docs/storage.md b/docs/storage.md index de8a8ed065..848c0d58e8 100644 --- a/docs/storage.md +++ b/docs/storage.md @@ -16,7 +16,7 @@ contains a chunks subdirectory with all the time series samples for that time wi a metadata file, and an index file (which maps metric names and labels to the time series in the chunks directory). By default, the samples in the chunks directory are organized into one or more segment files, each up to 512 MB. When series are deleted via the API, -deletion records are stored in separate tombstone files rather than being immediately +deletion records are stored in separate tombstone files rather than being immediately removed from the chunk segments. The current block for incoming samples is kept in memory and is not fully From 407b697ee2986430891a56c439167bf25113ad9b Mon Sep 17 00:00:00 2001 From: Raul Leite Date: Sat, 15 Nov 2025 10:55:26 -0600 Subject: [PATCH 3/3] structure adjusted as reccomended Corrected the structure of the explanation regarding how samples are organized in the chunks directory and the handling of deletion records. Signed-off-by: Raul Leite --- docs/storage.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/docs/storage.md b/docs/storage.md index 848c0d58e8..1feb0a4940 100644 --- a/docs/storage.md +++ b/docs/storage.md @@ -14,9 +14,9 @@ Prometheus's local time series database stores data in a custom, highly efficien Ingested samples are grouped into two-hour blocks. Each block consists of a directory that contains a chunks subdirectory with all the time series samples for that time window, a metadata file, and an index file (which maps metric names and labels to the time series -in the chunks directory). By default, the samples in the chunks directory are organized -into one or more segment files, each up to 512 MB. When series are deleted via the API, -deletion records are stored in separate tombstone files rather than being immediately +in the chunks directory). The samples in the chunks directory are organized into one +or more segment files, each up to 512 MB by default. When series are deleted via the API, +the deletion records are stored in separate tombstone files rather than being immediately removed from the chunk segments. The current block for incoming samples is kept in memory and is not fully