From f01a9ab3e749f4c26abde90d33eab76a64a5c6e8 Mon Sep 17 00:00:00 2001 From: meelahme Date: Tue, 25 Feb 2025 17:17:36 -0800 Subject: [PATCH 01/21] Adding a no_sync documentation to get-started page (#5826) --- content/shared/v3-core-get-started/_index.md | 7 +++++++ 1 file changed, 7 insertions(+) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index c216759bb..ae99683e8 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -323,6 +323,13 @@ For more information, see [diskless architecture](#diskless-architecture). > Because InfluxDB sends a write response after the WAL file has been flushed to the configured object store (default is every second), individual write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. > Future enhancements will include an API parameter that lets requests return without waiting for the WAL flush. +##### no_sync Write Option + +InfluxDB provides a no_sync write option to allow faster response of write request by skipping the wait for WAL presistence. When `no_sync=true`, InfluxDB writes data to WAL but immediately acknowledges the write request without waiting for persistence. + +- Default behavior(`no_sync=false`): Ensures data is persisted before acknowledging writes. Thus, reducing the risk of data loss. +- With `no_sync=true`: Improves write performance but increases the risk of data loss in case of crashes before WAL persistence. + #### Create a database or Table To create a database without writing data, use the `create` subcommand--for example: From d6109b3b0eed878a15bf47ea61b40b0427080d4a Mon Sep 17 00:00:00 2001 From: meelahme Date: Tue, 25 Feb 2025 18:42:10 -0800 Subject: [PATCH 02/21] adding CLI example to no_sync write option --- content/shared/v3-core-get-started/_index.md | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 81c905720..6ce8261fd 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -327,15 +327,20 @@ For more information, see [diskless architecture](#diskless-architecture). > Because InfluxDB sends a write response after the WAL file has been flushed to the configured object store (default is every second), individual write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. > Future enhancements will include an API parameter that lets requests return without waiting for the WAL flush. +##### No sync write option -##### no_sync Write Option - -InfluxDB provides a no_sync write option to allow faster response of write request by skipping the wait for WAL presistence. When `no_sync=true`, InfluxDB writes data to WAL but immediately acknowledges the write request without waiting for persistence. +InfluxDB provides a `no_sync` write option to allow faster response of write request by skipping the wait for WAL presistence. When `no_sync=true`, InfluxDB writes data to WAL but immediately acknowledges the write request without waiting for persistence. - Default behavior(`no_sync=false`): Ensures data is persisted before acknowledging writes. Thus, reducing the risk of data loss. - With `no_sync=true`: Improves write performance but increases the risk of data loss in case of crashes before WAL persistence. -#### Create a database or table +If you are using the CLI, here is an example of how to enable the `no_sync` option: + +```sh +influx write --bucket=mydb --org=my_org --token=my-token --no-sync +``` + +#### Create a database or Table To create a database without writing data, use the `create` subcommand--for example: From 716bedceb79f890843d8035ccf18003c4728b8f5 Mon Sep 17 00:00:00 2001 From: meelahme Date: Tue, 25 Feb 2025 19:00:59 -0800 Subject: [PATCH 03/21] Adding HTTP documentation for no_sync option --- content/shared/v3-core-get-started/_index.md | 11 ++++++++++- 1 file changed, 10 insertions(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 6ce8261fd..763e94baa 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -334,11 +334,20 @@ InfluxDB provides a `no_sync` write option to allow faster response of write req - Default behavior(`no_sync=false`): Ensures data is persisted before acknowledging writes. Thus, reducing the risk of data loss. - With `no_sync=true`: Improves write performance but increases the risk of data loss in case of crashes before WAL persistence. +If you are using the HTTP API, here is an example of how to enable the `no_sync` option: + +```sh +curl -v "http://localhost:8181/api/v3/write_lp?db=sensors&precision=auto&no_sync=true" \ + --data-raw "home,room=Sunroom temp=96 +home,room=Sunroom temp=hi" +``` + If you are using the CLI, here is an example of how to enable the `no_sync` option: ```sh -influx write --bucket=mydb --org=my_org --token=my-token --no-sync +influxdb3 write --bucket=mydb --org=my_org --token=my-token --no-sync ``` +Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. #### Create a database or Table From 329d12c2334131eba654846093834eaf41a7eaf9 Mon Sep 17 00:00:00 2001 From: meelahme Date: Wed, 26 Feb 2025 08:32:37 -0800 Subject: [PATCH 04/21] Updates to no_sync example for HTTP --- content/shared/v3-core-get-started/_index.md | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 763e94baa..335d55e85 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -338,8 +338,7 @@ If you are using the HTTP API, here is an example of how to enable the `no_sync` ```sh curl -v "http://localhost:8181/api/v3/write_lp?db=sensors&precision=auto&no_sync=true" \ - --data-raw "home,room=Sunroom temp=96 -home,room=Sunroom temp=hi" + --data-raw "home,room=Sunroom temp=96" ``` If you are using the CLI, here is an example of how to enable the `no_sync` option: From b89b6b0528b21dd7e5225504302738d2122ed895 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Thu, 27 Feb 2025 17:00:09 -0800 Subject: [PATCH 05/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 1 - 1 file changed, 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 335d55e85..d4a8b91ec 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -325,7 +325,6 @@ For more information, see [diskless architecture](#diskless-architecture). > ##### Write requests return after WAL flush > > Because InfluxDB sends a write response after the WAL file has been flushed to the configured object store (default is every second), individual write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. -> Future enhancements will include an API parameter that lets requests return without waiting for the WAL flush. ##### No sync write option From 7713957f68fa7d5f573ad6bd48f55bf821d55a3d Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Thu, 27 Feb 2025 17:00:17 -0800 Subject: [PATCH 06/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index d4a8b91ec..aab5f5bd6 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -328,7 +328,9 @@ For more information, see [diskless architecture](#diskless-architecture). ##### No sync write option -InfluxDB provides a `no_sync` write option to allow faster response of write request by skipping the wait for WAL presistence. When `no_sync=true`, InfluxDB writes data to WAL but immediately acknowledges the write request without waiting for persistence. +InfluxDB provides a `no_sync` write option that allows the write request to respond faster by skipping the wait for WAL presistence. When `no_sync=true`, InfluxDB writes data to the WAL and then immediately acknowledges the write request without waiting for persistence to the Object store. + +Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. - Default behavior(`no_sync=false`): Ensures data is persisted before acknowledging writes. Thus, reducing the risk of data loss. - With `no_sync=true`: Improves write performance but increases the risk of data loss in case of crashes before WAL persistence. From e7f4bc2ad62bb61761f688b530d69ef52623e5df Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Thu, 27 Feb 2025 17:00:25 -0800 Subject: [PATCH 07/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index aab5f5bd6..1850cc20b 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -332,7 +332,7 @@ InfluxDB provides a `no_sync` write option that allows the write request to resp Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. -- Default behavior(`no_sync=false`): Ensures data is persisted before acknowledging writes. Thus, reducing the risk of data loss. +- Default behavior(`no_sync=false`): Waits for data to be written to disk before acknowledging writes. This reduces the risk of data loss but increases latency for writes. - With `no_sync=true`: Improves write performance but increases the risk of data loss in case of crashes before WAL persistence. If you are using the HTTP API, here is an example of how to enable the `no_sync` option: From fa9a1bf10341c7cb1114789bb8b5225ec8ed8fb0 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Thu, 27 Feb 2025 17:00:34 -0800 Subject: [PATCH 08/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 1850cc20b..71b732561 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -333,7 +333,7 @@ InfluxDB provides a `no_sync` write option that allows the write request to resp Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. - Default behavior(`no_sync=false`): Waits for data to be written to disk before acknowledging writes. This reduces the risk of data loss but increases latency for writes. -- With `no_sync=true`: Improves write performance but increases the risk of data loss in case of crashes before WAL persistence. +- With `no_sync=true`: Reduces write latency but increases the risk of data loss in case of crashes before WAL persistence. If you are using the HTTP API, here is an example of how to enable the `no_sync` option: From d0af3b3742077935f068c165c0e1df9059c6b7be Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Thu, 27 Feb 2025 17:01:01 -0800 Subject: [PATCH 09/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 71b732561..b5007f949 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -335,7 +335,10 @@ Using `no_sync=true` is best when prioritizing high-throughput writes over absol - Default behavior(`no_sync=false`): Waits for data to be written to disk before acknowledging writes. This reduces the risk of data loss but increases latency for writes. - With `no_sync=true`: Reduces write latency but increases the risk of data loss in case of crashes before WAL persistence. -If you are using the HTTP API, here is an example of how to enable the `no_sync` option: +###### Immediate write using the HTTP API + +The `no_sync` parameter controls when writes are acknowledged--for example: + ```sh curl -v "http://localhost:8181/api/v3/write_lp?db=sensors&precision=auto&no_sync=true" \ From 826c98787f36dbcec676ee26b0b373bb34bf95a4 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Thu, 27 Feb 2025 17:01:23 -0800 Subject: [PATCH 10/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 1 - 1 file changed, 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index b5007f949..3e0514eca 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -350,7 +350,6 @@ If you are using the CLI, here is an example of how to enable the `no_sync` opti ```sh influxdb3 write --bucket=mydb --org=my_org --token=my-token --no-sync ``` -Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. #### Create a database or Table From 8c5aeb024015105a6cca5825055b5d6c92e4adf6 Mon Sep 17 00:00:00 2001 From: meelahme Date: Thu, 27 Feb 2025 18:18:23 -0800 Subject: [PATCH 11/21] Reverting sentence delete to note section --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 3e0514eca..b130df955 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -324,7 +324,7 @@ For more information, see [diskless architecture](#diskless-architecture). > [!Note] > ##### Write requests return after WAL flush > -> Because InfluxDB sends a write response after the WAL file has been flushed to the configured object store (default is every second), individual write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. +> Because InfluxDB sends a write response after the WAL file has been flushed to the configured object store (default is every second), individual write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. Future enhancements will include an API parameter that lets requests return without waiting for the WAL flush. ##### No sync write option From b04c7ef9bd5c8eeb807f9ac32db3c9b9abc64a62 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Fri, 28 Feb 2025 10:14:54 -0800 Subject: [PATCH 12/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index b130df955..ef5907c16 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -351,7 +351,7 @@ If you are using the CLI, here is an example of how to enable the `no_sync` opti influxdb3 write --bucket=mydb --org=my_org --token=my-token --no-sync ``` -#### Create a database or Table +#### Create a database or table To create a database without writing data, use the `create` subcommand--for example: From 9c00e4b179dfc178b06784a46f1d40f1c07a42a8 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Fri, 28 Feb 2025 10:16:23 -0800 Subject: [PATCH 13/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index ef5907c16..4d0b1e75a 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -324,7 +324,8 @@ For more information, see [diskless architecture](#diskless-architecture). > [!Note] > ##### Write requests return after WAL flush > -> Because InfluxDB sends a write response after the WAL file has been flushed to the configured object store (default is every second), individual write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. Future enhancements will include an API parameter that lets requests return without waiting for the WAL flush. +> By default, when you write data, InfluxDB sends a write response after the WAL file has been flushed to the Object store (default is every second). Write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. +For faster write responses, you can use the [`no_sync` option](#no-sync-write-option) to skip the wait for WAL persistence. ##### No sync write option From c28c8e91b958c8b6827b729c2b7fc6c437a20ff5 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Fri, 28 Feb 2025 10:19:29 -0800 Subject: [PATCH 14/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 4d0b1e75a..bd1bc9d71 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -329,7 +329,7 @@ For faster write responses, you can use the [`no_sync` option](#no-sync-write-op ##### No sync write option -InfluxDB provides a `no_sync` write option that allows the write request to respond faster by skipping the wait for WAL presistence. When `no_sync=true`, InfluxDB writes data to the WAL and then immediately acknowledges the write request without waiting for persistence to the Object store. +InfluxDB provides a `no_sync` write option that allows the write request to respond faster by skipping the wait for WAL persistence. When `no_sync=true`, InfluxDB writes data to the WAL and then immediately acknowledges the write request without waiting for persistence to the Object store. Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. From 0fc05b651d71d69e0167a5f235b9f4f66547d1ef Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Fri, 28 Feb 2025 10:19:57 -0800 Subject: [PATCH 15/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index bd1bc9d71..ba5e6250a 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -333,7 +333,7 @@ InfluxDB provides a `no_sync` write option that allows the write request to resp Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. -- Default behavior(`no_sync=false`): Waits for data to be written to disk before acknowledging writes. This reduces the risk of data loss but increases latency for writes. +- Default behavior (`no_sync=false`): Waits for data to be written to disk before acknowledging writes. This reduces the risk of data loss but increases latency for writes. - With `no_sync=true`: Reduces write latency but increases the risk of data loss in case of crashes before WAL persistence. ###### Immediate write using the HTTP API From a5b8c4488170a81a0e349c7bdf5c145ca4eb4ee3 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Fri, 28 Feb 2025 10:20:22 -0800 Subject: [PATCH 16/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index ba5e6250a..9a6c09a6e 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -342,7 +342,7 @@ The `no_sync` parameter controls when writes are acknowledged--for example: ```sh -curl -v "http://localhost:8181/api/v3/write_lp?db=sensors&precision=auto&no_sync=true" \ +curl "http://localhost:8181/api/v3/write_lp?db=sensors&precision=auto&no_sync=true" \ --data-raw "home,room=Sunroom temp=96" ``` From 95a88f6d90e1def01e19a8d86cce5ac7e00e9d67 Mon Sep 17 00:00:00 2001 From: Jameelah Mercer <36314199+MeelahMe@users.noreply.github.com> Date: Fri, 28 Feb 2025 10:20:40 -0800 Subject: [PATCH 17/21] Update content/shared/v3-core-get-started/_index.md Co-authored-by: Jason Stirnaman --- content/shared/v3-core-get-started/_index.md | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 9a6c09a6e..681c59ccb 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -346,7 +346,9 @@ curl "http://localhost:8181/api/v3/write_lp?db=sensors&precision=auto&no_sync=tr --data-raw "home,room=Sunroom temp=96" ``` -If you are using the CLI, here is an example of how to enable the `no_sync` option: +###### Immediate write using the influxdb3 CLI + +The `no_sync` CLI option controls when writes are acknowledged--for example: ```sh influxdb3 write --bucket=mydb --org=my_org --token=my-token --no-sync From 6448be2c74d25c6ba9eab40fe38bdc9f08ede7be Mon Sep 17 00:00:00 2001 From: Jason Stirnaman Date: Wed, 5 Mar 2025 16:54:17 -0600 Subject: [PATCH 18/21] Update content/shared/v3-core-get-started/_index.md --- content/shared/v3-core-get-started/_index.md | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 180b4ff61..d2c1d8ce3 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -329,8 +329,9 @@ For more information, see [diskless architecture](#diskless-architecture). > [!Note] > ##### Write requests return after WAL flush > -> By default, when you write data, InfluxDB sends a write response after the WAL file has been flushed to the Object store (default is every second). Write requests might not complete quickly, but you can make many concurrent requests to achieve higher total throughput. -For faster write responses, you can use the [`no_sync` option](#no-sync-write-option) to skip the wait for WAL persistence. +> By default, InfluxDB acknowledges writes after flushing the WAL file to the Object store (occurring every second). For high throughput, you can send multiple concurrent write requests. +> +> To reduce the latency of writes, use the [`no_sync` write option](#no-sync-write-option), which acknowledges writes _before_ WAL persistence completes. ##### No sync write option From 9ce8d93d46b0473c0d3d7072b11902e3186c7a9d Mon Sep 17 00:00:00 2001 From: Jason Stirnaman Date: Wed, 5 Mar 2025 16:54:34 -0600 Subject: [PATCH 19/21] Update content/shared/v3-core-get-started/_index.md --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index d2c1d8ce3..0097034c3 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -335,7 +335,7 @@ For more information, see [diskless architecture](#diskless-architecture). ##### No sync write option -InfluxDB provides a `no_sync` write option that allows the write request to respond faster by skipping the wait for WAL persistence. When `no_sync=true`, InfluxDB writes data to the WAL and then immediately acknowledges the write request without waiting for persistence to the Object store. +The `no_sync` write option reduces latency by acknowledging write requests before WAL persistence completes. When set to `true`, InfluxDB validates the data, writes the data to the WAL, and then immediately confirms the write, without waiting for persistence to the Object store. Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. From 9c276ee8c9f5c644245ed604799900be2f603dcb Mon Sep 17 00:00:00 2001 From: Jason Stirnaman Date: Wed, 5 Mar 2025 16:54:48 -0600 Subject: [PATCH 20/21] Update content/shared/v3-core-get-started/_index.md --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 0097034c3..5441b04ec 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -339,7 +339,7 @@ The `no_sync` write option reduces latency by acknowledging write requests befor Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. -- Default behavior (`no_sync=false`): Waits for data to be written to disk before acknowledging writes. This reduces the risk of data loss but increases latency for writes. +- Default behavior (`no_sync=false`): Waits for data to be written to the Object store before acknowledging the write. Reduces the risk of data loss, but increases the latency of the response. - With `no_sync=true`: Reduces write latency but increases the risk of data loss in case of crashes before WAL persistence. ###### Immediate write using the HTTP API From 1d6423730857bc644b2ae19af7e65c116ae0810d Mon Sep 17 00:00:00 2001 From: Jason Stirnaman Date: Wed, 5 Mar 2025 16:54:55 -0600 Subject: [PATCH 21/21] Update content/shared/v3-core-get-started/_index.md --- content/shared/v3-core-get-started/_index.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/content/shared/v3-core-get-started/_index.md b/content/shared/v3-core-get-started/_index.md index 5441b04ec..b563a4b4d 100644 --- a/content/shared/v3-core-get-started/_index.md +++ b/content/shared/v3-core-get-started/_index.md @@ -340,7 +340,7 @@ The `no_sync` write option reduces latency by acknowledging write requests befor Using `no_sync=true` is best when prioritizing high-throughput writes over absolute durability. - Default behavior (`no_sync=false`): Waits for data to be written to the Object store before acknowledging the write. Reduces the risk of data loss, but increases the latency of the response. -- With `no_sync=true`: Reduces write latency but increases the risk of data loss in case of crashes before WAL persistence. +- With `no_sync=true`: Reduces write latency, but increases the risk of data loss in case of a crash before WAL persistence. ###### Immediate write using the HTTP API