[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[libeufin] 02/02: comment
From: |
gnunet |
Subject: |
[libeufin] 02/02: comment |
Date: |
Tue, 11 Apr 2023 11:02:19 +0200 |
This is an automated email from the git hooks/post-receive script.
ms pushed a commit to branch master
in repository libeufin.
commit 1b9fbb02ac2d2eb402c18877079d1bdbae03a620
Author: MS <ms@taler.net>
AuthorDate: Sun Apr 9 18:25:02 2023 +0200
comment
---
nexus/src/test/kotlin/SchedulingTest.kt | 11 +++++++++++
1 file changed, 11 insertions(+)
diff --git a/nexus/src/test/kotlin/SchedulingTest.kt
b/nexus/src/test/kotlin/SchedulingTest.kt
index 4cb54390..0e1b4be9 100644
--- a/nexus/src/test/kotlin/SchedulingTest.kt
+++ b/nexus/src/test/kotlin/SchedulingTest.kt
@@ -21,6 +21,17 @@ import kotlin.text.get
* give the chance to monitor the CPU usage with TOP(1)
*/
+/**
+ * It emerged that whether asking transactions via EBICS or x-libeufin-bank
+ * is NOT performance relevant! For example, asking for a bank account
+ * balance - via the plain Access API - brings the CPU usage to > 10%. Asking
+ * for /config - via Integration API - used to oscillate the CPU usage
+ * between 3 and 10%.
+ *
+ * The scheduler's loop style is not relevant either: a while-true &
delay(1000)
+ * or a Java Timer did NOT change the perf.
+ */
+
// This class focuses on the perf. of Nexus scheduling.
class SchedulingTest {
// Launching the scheduler to measure its perf with TOP(1)
--
To stop receiving notification emails like this one, please contact
gnunet@gnunet.org.