This is an automated email from the ASF dual-hosted git repository.
ethanfeng pushed a commit to branch main
in repository https://gitbox.apache.org/repos/asf/incubator-celeborn.git
The following commit(s) were added to refs/heads/main by this push:
new 277f7ced5 [CELEBORN-1187] Unify the size and file count of active
shuffle metrics for master and worker
277f7ced5 is described below
commit 277f7ced5787e904579e4159ebcb9bc6cb9d4367
Author: SteNicholas <[email protected]>
AuthorDate: Fri Dec 22 17:07:39 2023 +0800
[CELEBORN-1187] Unify the size and file count of active shuffle metrics for
master and worker
### What changes were proposed in this pull request?
Unify the size and file count of active shuffle metrics for `MasterSource`
and `WorkerSource`.
### Why are the changes needed?
`MasterSource` uses `PartitionWritten` and `PartitionFileCount` metrics as
the size and file count of active shuffle for all workers. Meanwhile,
`WorkerSource` uses `ActiveShuffleSize` and `ActiveShuffleFileCount` metrics as
the size and file count of active shuffle for a worker including master replica
and slave replica. It's recommended to unify the size and file count of active
shuffle metrics between `MasterSource` and `WorkerSource`.
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Internal tests.
Closes #2171 from SteNicholas/CELEBORN-1187.
Lead-authored-by: SteNicholas <[email protected]>
Co-authored-by: 蒋晓峰 <[email protected]>
Signed-off-by: mingji <[email protected]>
---
METRICS.md | 8 +++-----
assets/grafana/celeborn-dashboard.json | 12 ++++++------
docs/migration.md | 4 ++++
docs/monitoring.md | 4 ++--
.../org/apache/celeborn/service/deploy/master/Master.scala | 4 ++--
.../apache/celeborn/service/deploy/master/MasterSource.scala | 4 ++--
6 files changed, 19 insertions(+), 17 deletions(-)
diff --git a/METRICS.md b/METRICS.md
index c4e0862d4..ec0ffad10 100644
--- a/METRICS.md
+++ b/METRICS.md
@@ -81,13 +81,13 @@ Here is an example of Grafana dashboard importing.
| RunningApplicationCount | master |
The count of running applications in the cluster.
|
| OfferSlotsTime | master |
The time of offer slots.
|
| PartitionSize | master | The
estimated partition size of last 20 flush window whose length is 15 seconds by
defaults. |
-| PartitionWritten | master |
The active shuffle size.
|
-| PartitionFileCount | master |
The active shuffle partition count.
|
+| RegisteredShuffleCount | master and worker |
The value means count of registered shuffle.
|
+| ActiveShuffleSize | master and worker | The value
means the active shuffle size for workers or a worker including master replica
and slave replica. |
+| ActiveShuffleFileCount | master and worker | The value
means the active shuffle size for workers or a worker including master replica
and slave replica. |
| diskFileCount | master and worker |
The count of disk files consumption by each user.
|
| diskBytesWritten | master and worker |
The amount of disk files consumption by each user.
|
| hdfsFileCount | master and worker |
The count of hdfs files consumption by each user.
|
| hdfsBytesWritten | master and worker |
The amount of hdfs files consumption by each user.
|
-| RegisteredShuffleCount | master and worker |
The value means count of registered shuffle.
|
| CommitFilesTime | worker |
CommitFiles means flush and close a shuffle partition file.
|
| ReserveSlotsTime | worker |
ReserveSlots means acquire a disk buffer and record partition location.
|
| FlushDataTime | worker |
FlushData means flush a disk buffer to disk.
|
@@ -113,8 +113,6 @@ Here is an example of Grafana dashboard importing.
| DiskBuffer | worker | Disk buffers
are part of netty used memory, means data need to write to disk but haven't
been written to disk. |
| PausePushData | worker |
PausePushData means the count of worker stopped receiving data from client.
|
| PausePushDataAndReplicate | worker |
PausePushDataAndReplicate means the count of worker stopped receiving data from
client and other workers. |
-| ActiveShuffleSize | worker |
The active shuffle size of a worker including master replica and slave replica.
|
-| ActiveShuffleFileCount | worker |
The active shuffle file count of a worker including master replica and slave
replica. |
| jvm_gc_count | JVM |
The GC count of each garbage collector.
|
| jvm_gc_time | JVM |
The GC cost time of each garbage collector.
|
| jvm_memory_heap_init | JVM |
The amount of heap init memory.
|
diff --git a/assets/grafana/celeborn-dashboard.json
b/assets/grafana/celeborn-dashboard.json
index e2a3dc30c..653d5863e 100644
--- a/assets/grafana/celeborn-dashboard.json
+++ b/assets/grafana/celeborn-dashboard.json
@@ -626,13 +626,13 @@
"uid": "${DS_PROMETHEUS}"
},
"editorMode": "code",
- "expr": "metrics_PartitionWritten_Value",
+ "expr": "metrics_ActiveShuffleSize_Value{role=\"Master\"}",
"legendFormat": "${baseLegend}",
"range": true,
"refId": "A"
}
],
- "title": "metrics_PartitionWritten_Value",
+ "title": "metrics_ActiveShuffleSize_Value",
"type": "timeseries"
},
{
@@ -718,13 +718,13 @@
"uid": "${DS_PROMETHEUS}"
},
"editorMode": "code",
- "expr": "metrics_PartitionFileCount_Value",
+ "expr": "metrics_ActiveShuffleFileCount_Value{role=\"Master\"}",
"legendFormat": "${baseLegend}",
"range": true,
"refId": "A"
}
],
- "title": "metrics_PartitionFileCount_Value",
+ "title": "metrics_ActiveShuffleFileCount_Value",
"type": "timeseries"
},
{
@@ -1845,7 +1845,7 @@
"uid": "${DS_PROMETHEUS}"
},
"editorMode": "builder",
- "expr": "metrics_ActiveShuffleSize_Value",
+ "expr": "metrics_ActiveShuffleSize_Value{role=\"Worker\"}",
"legendFormat": "${baseLegend}",
"instant": false,
"range": true,
@@ -1964,7 +1964,7 @@
"uid": "${DS_PROMETHEUS}"
},
"editorMode": "builder",
- "expr": "metrics_ActiveShuffleFileCount_Value",
+ "expr": "metrics_ActiveShuffleFileCount_Value{role=\"Worker\"}",
"legendFormat": "${baseLegend}",
"instant": false,
"range": true,
diff --git a/docs/migration.md b/docs/migration.md
index 15937d3e9..538b7ada0 100644
--- a/docs/migration.md
+++ b/docs/migration.md
@@ -44,6 +44,10 @@ license: |
- Since 0.4.0, Celeborn deprecate `celeborn.storage.activeTypes`. Please use
`celeborn.storage.availableTypes` instead.
+- Since 0.4.0, Celeborn master metrics `PartitionWritten` is renamed as
`ActiveShuffleSize`.
+
+- Since 0.4.0, Celeborn master metrics `PartitionFileCount` is renamed as
`ActiveShuffleFileCount`.
+
## Upgrading from 0.3.1 to 0.3.2
- Since 0.3.1, Celeborn changed the default value of
`raft.client.rpc.request.timeout` from `3s` to `10s`.
diff --git a/docs/monitoring.md b/docs/monitoring.md
index 3aa210907..64592932f 100644
--- a/docs/monitoring.md
+++ b/docs/monitoring.md
@@ -99,9 +99,9 @@ These metrics are exposed by Celeborn master.
- IsActiveMaster
- PartitionSize
- The size of estimated shuffle partition.
- - PartitionWritten
+ - ActiveShuffleSize
- The active shuffle size of workers.
- - PartitionFileCount
+ - ActiveShuffleFileCount
- The active shuffle file count of workers.
- OfferSlotsTime
- The time for masters to handle `RequestSlots` request when
registering shuffle.
diff --git
a/master/src/main/scala/org/apache/celeborn/service/deploy/master/Master.scala
b/master/src/main/scala/org/apache/celeborn/service/deploy/master/Master.scala
index 2235d8cfc..3fcb94b01 100644
---
a/master/src/main/scala/org/apache/celeborn/service/deploy/master/Master.scala
+++
b/master/src/main/scala/org/apache/celeborn/service/deploy/master/Master.scala
@@ -168,7 +168,7 @@ private[celeborn] class Master(
statusSystem.appHeartbeatTime.size
}
masterSource.addGauge(MasterSource.PARTITION_SIZE) { () =>
statusSystem.estimatedPartitionSize }
- masterSource.addGauge(MasterSource.PARTITION_WRITTEN) { () =>
+ masterSource.addGauge(MasterSource.ACTIVE_SHUFFLE_SIZE) { () =>
statusSystem.workers.parallelStream()
.mapToLong(new ToLongFunction[WorkerInfo]() {
override def applyAsLong(value: WorkerInfo): Long =
@@ -178,7 +178,7 @@ private[celeborn] class Master(
}).sum()
}).sum()
}
- masterSource.addGauge(MasterSource.PARTITION_FILE_COUNT) { () =>
+ masterSource.addGauge(MasterSource.ACTIVE_SHUFFLE_FILE_COUNT) { () =>
statusSystem.workers.parallelStream()
.mapToLong(new ToLongFunction[WorkerInfo]() {
override def applyAsLong(value: WorkerInfo): Long =
diff --git
a/master/src/main/scala/org/apache/celeborn/service/deploy/master/MasterSource.scala
b/master/src/main/scala/org/apache/celeborn/service/deploy/master/MasterSource.scala
index 8010f0a74..0fbb35121 100644
---
a/master/src/main/scala/org/apache/celeborn/service/deploy/master/MasterSource.scala
+++
b/master/src/main/scala/org/apache/celeborn/service/deploy/master/MasterSource.scala
@@ -46,9 +46,9 @@ object MasterSource {
val PARTITION_SIZE = "PartitionSize"
- val PARTITION_WRITTEN = "PartitionWritten"
+ val ACTIVE_SHUFFLE_SIZE = "ActiveShuffleSize"
- val PARTITION_FILE_COUNT = "PartitionFileCount"
+ val ACTIVE_SHUFFLE_FILE_COUNT = "ActiveShuffleFileCount"
val OFFER_SLOTS_TIME = "OfferSlotsTime"
}
