This is an automated email from the ASF dual-hosted git repository.
roryqi pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-uniffle.git
The following commit(s) were added to refs/heads/master by this push:
new 72d85a778 [#1472][part-4] feature(server): Add metrics for Netty's
pinnedDirectMemory and usedDirectMemory (#1524)
72d85a778 is described below
commit 72d85a778f559ddf1efe0b6909eb8146e5a1c646
Author: RickyMa <[email protected]>
AuthorDate: Wed Feb 14 10:23:25 2024 +0800
[#1472][part-4] feature(server): Add metrics for Netty's pinnedDirectMemory
and usedDirectMemory (#1524)
### What changes were proposed in this pull request?
We need to know the exact direct memory usage of `PooledByteBufAllocator`
in `Netty`.
So we should introduce metrics for Netty's `pinnedDirectMemory` and
`usedDirectMemory`.
### Why are the changes needed?
A sub PR for: https://github.com/apache/incubator-uniffle/pull/1519
### Does this PR introduce _any_ user-facing change?
No.
### How was this patch tested?
Existing UTs.
---
.../org/apache/uniffle/server/NettyDirectMemoryTracker.java | 12 +++++++++++-
.../java/org/apache/uniffle/server/ShuffleServerMetrics.java | 6 ++++++
2 files changed, 17 insertions(+), 1 deletion(-)
diff --git
a/server/src/main/java/org/apache/uniffle/server/NettyDirectMemoryTracker.java
b/server/src/main/java/org/apache/uniffle/server/NettyDirectMemoryTracker.java
index 208e41595..c3a31fd52 100644
---
a/server/src/main/java/org/apache/uniffle/server/NettyDirectMemoryTracker.java
+++
b/server/src/main/java/org/apache/uniffle/server/NettyDirectMemoryTracker.java
@@ -25,6 +25,7 @@ import io.netty.util.internal.PlatformDependent;
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
+import org.apache.uniffle.common.util.NettyUtils;
import org.apache.uniffle.common.util.ThreadUtils;
public class NettyDirectMemoryTracker {
@@ -54,10 +55,19 @@ public class NettyDirectMemoryTracker {
() -> {
try {
long usedDirectMemory = PlatformDependent.usedDirectMemory();
+ long allocatedDirectMemory =
+
NettyUtils.getNettyBufferAllocator().metric().usedDirectMemory();
+ long pinnedDirectMemory =
NettyUtils.getNettyBufferAllocator().pinnedDirectMemory();
if (LOG.isDebugEnabled()) {
- LOG.debug("Current direct memory usage: {}", usedDirectMemory);
+ LOG.debug(
+ "Current usedDirectMemory:{}, allocatedDirectMemory:{},
pinnedDirectMemory:{}",
+ usedDirectMemory,
+ allocatedDirectMemory,
+ pinnedDirectMemory);
}
ShuffleServerMetrics.gaugeUsedDirectMemorySize.set(usedDirectMemory);
+
ShuffleServerMetrics.gaugeAllocatedDirectMemorySize.set(allocatedDirectMemory);
+
ShuffleServerMetrics.gaugePinnedDirectMemorySize.set(pinnedDirectMemory);
} catch (Throwable t) {
LOG.error("Failed to report direct memory.", t);
}
diff --git
a/server/src/main/java/org/apache/uniffle/server/ShuffleServerMetrics.java
b/server/src/main/java/org/apache/uniffle/server/ShuffleServerMetrics.java
index c5aeac615..649e5044b 100644
--- a/server/src/main/java/org/apache/uniffle/server/ShuffleServerMetrics.java
+++ b/server/src/main/java/org/apache/uniffle/server/ShuffleServerMetrics.java
@@ -79,6 +79,8 @@ public class ShuffleServerMetrics {
private static final String USED_BUFFER_SIZE = "used_buffer_size";
private static final String READ_USED_BUFFER_SIZE = "read_used_buffer_size";
private static final String USED_DIRECT_MEMORY_SIZE =
"used_direct_memory_size";
+ private static final String ALLOCATED_DIRECT_MEMORY_SIZE =
"allocated_direct_memory_size";
+ private static final String PINNED_DIRECT_MEMORY_SIZE =
"pinned_direct_memory_size";
private static final String TOTAL_FAILED_WRITTEN_EVENT_NUM =
"total_failed_written_event_num";
private static final String TOTAL_DROPPED_EVENT_NUM =
"total_dropped_event_num";
private static final String TOTAL_HADOOP_WRITE_DATA =
"total_hadoop_write_data";
@@ -184,6 +186,8 @@ public class ShuffleServerMetrics {
public static Gauge.Child gaugeUsedBufferSize;
public static Gauge.Child gaugeReadBufferUsedSize;
public static Gauge.Child gaugeUsedDirectMemorySize;
+ public static Gauge.Child gaugeAllocatedDirectMemorySize;
+ public static Gauge.Child gaugePinnedDirectMemorySize;
public static Gauge.Child gaugeWriteHandler;
public static Gauge.Child gaugeEventQueueSize;
public static Gauge.Child gaugeHadoopFlushThreadPoolQueueSize;
@@ -380,6 +384,8 @@ public class ShuffleServerMetrics {
gaugeUsedBufferSize = metricsManager.addLabeledGauge(USED_BUFFER_SIZE);
gaugeReadBufferUsedSize =
metricsManager.addLabeledGauge(READ_USED_BUFFER_SIZE);
gaugeUsedDirectMemorySize =
metricsManager.addLabeledGauge(USED_DIRECT_MEMORY_SIZE);
+ gaugeAllocatedDirectMemorySize =
metricsManager.addLabeledGauge(ALLOCATED_DIRECT_MEMORY_SIZE);
+ gaugePinnedDirectMemorySize =
metricsManager.addLabeledGauge(PINNED_DIRECT_MEMORY_SIZE);
gaugeWriteHandler = metricsManager.addLabeledGauge(TOTAL_WRITE_HANDLER);
gaugeEventQueueSize = metricsManager.addLabeledGauge(EVENT_QUEUE_SIZE);
gaugeHadoopFlushThreadPoolQueueSize =