Github user ppadma commented on a diff in the pull request:
https://github.com/apache/drill/pull/928#discussion_r136376799
--- Diff:
exec/java-exec/src/main/java/org/apache/drill/exec/ExecConstants.java ---
@@ -373,13 +373,26 @@
String AVERAGE_FIELD_WIDTH_KEY = "planner.memory.average_field_width";
OptionValidator AVERAGE_FIELD_WIDTH = new
PositiveLongValidator(AVERAGE_FIELD_WIDTH_KEY, Long.MAX_VALUE);
+ // Resource management boot-time options.
+
+ String MAX_MEMORY_PER_NODE = "drill.exec.rm.memory_per_node";
+ String MAX_CPUS_PER_NODE = "drill.exec.rm.cpus_per_node";
+
+ // Resource management system run-time options.
+
+ // Enables queues. When running embedded, enables an in-process queue.
When
+ // running distributed, enables the Zookeeper-based distributed queue.
+
BooleanValidator ENABLE_QUEUE = new
BooleanValidator("exec.queue.enable");
LongValidator LARGE_QUEUE_SIZE = new
PositiveLongValidator("exec.queue.large", 1000);
LongValidator SMALL_QUEUE_SIZE = new
PositiveLongValidator("exec.queue.small", 100000);
- LongValidator QUEUE_THRESHOLD_SIZE = new
PositiveLongValidator("exec.queue.threshold",
- Long.MAX_VALUE);
- LongValidator QUEUE_TIMEOUT = new
PositiveLongValidator("exec.queue.timeout_millis",
- Long.MAX_VALUE);
+ LongValidator QUEUE_THRESHOLD_SIZE = new
PositiveLongValidator("exec.queue.threshold", Long.MAX_VALUE);
+ LongValidator QUEUE_TIMEOUT = new
PositiveLongValidator("exec.queue.timeout_millis", Long.MAX_VALUE);
+
+ // Ratio of memory for small queries vs. large queries.
+ // Each small query gets 1 unit, each large query gets
QUEUE_MEMORY_RATIO units.
+
+ DoubleValidator QUEUE_MEMORY_RATIO = new
RangeDoubleValidator("exec.queue.memory_ratio", 0.001, 1000);
--- End diff --
why lower bound of 0.001 ? should it be 1 ? I am thinking larger query gets
at least as much as smaller query.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
