Issue | Fix Version | Assignee | Priority | Status | Resolution | Resolved In Build |
---|---|---|---|---|---|---|
JDK-8208331 | 12 | Gustavo Romero | P3 | Resolved | Fixed | b05 |
JDK-8208428 | 11.0.2 | Gustavo Romero | P3 | Resolved | Fixed | b01 |
JDK-8209253 | 11.0.1 | Gustavo Romero | P3 | Resolved | Fixed | b05 |
UseNUMA does not check that node memory is available to the JVM processes before spreading the Java heap across regions for each node.
If the JVM isn't allowed to use memory on all of the nodes (by numactl, cgroups, docker, etc), then a significant fraction of the Java heap will be unusable, causing early GC.
This was seen on linux aarch64, using ParallelGC, but this doesn't look aarch64-specific. This is in JDK 9 /10 (probably always).
If the JVM isn't allowed to use memory on all of the nodes (by numactl, cgroups, docker, etc), then a significant fraction of the Java heap will be unusable, causing early GC.
This was seen on linux aarch64, using ParallelGC, but this doesn't look aarch64-specific. This is in JDK 9 /10 (probably always).
- backported by
-
JDK-8208331 UseNUMA memory interleaving vs membind
-
- Resolved
-
-
JDK-8208428 UseNUMA memory interleaving vs membind
-
- Resolved
-
-
JDK-8209253 UseNUMA memory interleaving vs membind
-
- Resolved
-
- relates to
-
JDK-8205051 Poor Performance with UseNUMA when cpu and memory nodes are misaligned
-
- Resolved
-
-
JDK-8213827 NUMA heap allocation does not respect process membind/interleave settings
-
- Resolved
-