Skip to content

[TESTING] Disable rate limiting in inference services #133067

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 1 commit into from
Closed

[TESTING] Disable rate limiting in inference services #133067

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
16 changes: 6 additions & 10 deletions ...n/java/org/elasticsearch/xpack/inference/external/http/sender/RequestExecutorService.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -449,27 +449,23 @@ public synchronized TimeValue executeEnqueuedTask() {
}

private TimeValue executeEnqueuedTaskInternal() {
var timeBeforeAvailableToken = rateLimiter.timeToReserve(1);
if (shouldExecuteImmediately(timeBeforeAvailableToken) == false) {
return timeBeforeAvailableToken;
}

var task = queue.poll();

// TODO Batching - in a situation where no new tasks are queued we'll want to execute any prepared tasks
// So we'll need to check for null and call a helper method executePreparedTasks()

if (shouldExecuteTask(task) == false) {
return NO_TASKS_AVAILABLE;
logger.warn(
"not executing task [{}] because it is null or has already completed",
task == null ? "null" : task.getRequestManager().inferenceEntityId()
);
return TimeValue.ZERO;
}

// We should never have to wait because we checked above
var reserveRes = rateLimiter.reserve(1);
assert shouldExecuteImmediately(reserveRes) : "Reserving request tokens required a sleep when it should not have";

task.getRequestManager()
.execute(task.getInferenceInputs(), requestSender, task.getRequestCompletedFunction(), task.getListener());
return EXECUTED_A_TASK;
return TimeValue.ZERO;
}

private static boolean shouldExecuteTask(RejectableTask task) {
Expand Down