aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorMaciek Konstantynowicz <mkonstan@cisco.com>2018-04-25 13:36:47 +0100
committerMaciek Konstantynowicz <mkonstan@cisco.com>2018-04-25 13:36:47 +0100
commit28f9f057aadad0b744dc91df3240caac7e311424 (patch)
treef7c9fbc037d290e291939b5e5148e73060e50788
parent4c6fe5602edcbd9857a846e5b13a21d5c671a2c8 (diff)
Edits of intro to perf trending dashboard
Change-Id: Ide9a38ac5798ba5a2c37f42232dc86b186502085 Signed-off-by: Maciek Konstantynowicz <mkonstan@cisco.com>
-rw-r--r--docs/cpta/introduction/index.rst45
1 files changed, 20 insertions, 25 deletions
diff --git a/docs/cpta/introduction/index.rst b/docs/cpta/introduction/index.rst
index 26fa3a19a9..af411c5632 100644
--- a/docs/cpta/introduction/index.rst
+++ b/docs/cpta/introduction/index.rst
@@ -4,35 +4,30 @@ VPP MRR Performance Dashboard
Description
-----------
-Dashboard tables list a summary of per test-case VPP MRR performance trend
-values and detected anomalies (Maximum Receive Rate - received packet rate
-under line rate load). Data comes from trending MRR jobs executed every 12
-hrs (2:00, 14:00 UTC). Trend, trend compliance and anomaly calculations are
-based on a rolling window of <N> samples, currently N=14 covering last 7 days.
-Separate tables are generated for tested VPP worker-thread-core combinations
-(1t1c, 2t2c, 4t4c).
+Dashboard tables list a summary of per test-case VPP MRR performance
+trend and trend compliance metrics, and detected number of anomalies.
+Data samples come from the CSIT VPP trending MRR jobs executed twice a
+day, every 12 hrs (02:00, 14:00 UTC). All trend and anomaly evaluation
+is based on a rolling window of <N=14> data samples, covering last 7
+days.
Legend to table:
- - **Test Case** : name of CSIT test case, see naming convention in
+ - **Test Case** : name of CSIT test case, naming convention in
`CSIT wiki <https://wiki.fd.io/view/CSIT/csit-test-naming>`_.
- - **Throughput Trend [Mpps]** : last value of trend calculated over a
- rolling window.
- - **Trend Compliance** : calculated based on detected anomalies over a
- rolling window, listed in precedence order - i) **failure** if 3
- consecutive outliers, ii) **regression** if any regressions, iii)
- **progression** if any progressions, iv) **normal** if data compliant
- with trend; test cases listed alphabetically within compliance category.
- - **Top Anomaly [Mpps]** : i) outlier if **failure**, ii) drop
- if **regression**, iii) gain if **progression**, iv) **-**
- if normal i.e. within trend.
- - **Change [%]** : **Top Anomaly** vs. **Throughput Trend**, **-** if
- normal.
- - **Outliers [Number]** : number of outliers detected over a rolling window.
-
-Tables are listed in sections 1.x. Followed by daily trending graphs in
-sections 2.x. Daily trending data used to generate the graphs is listed in
-sections 3.x.
+ - **Trend [Mpps]** : last value of trend.
+ - **Short-Term Change [%]** : Relative change of last trend value vs. last week trend value.
+ - **Long-Term Change [%]** : Relative change of last trend value vs. maximum of trend values over the last quarter except last week.
+ - **Regressions [#]** : Number of regressions detected.
+ - **Progressions [#]** : Number of progressions detected.
+ - **Outliers [#]** : Number of outliers detected.
+
+All trend and anomaly calculations are defined in :ref:`trending_methodology`.
+
+Tested VPP worker-thread-core combinations (1t1c, 2t2c, 4t4c) are listed
+in separate tables in section 1.x. Followed by trending methodology in
+section 2. and daily trending graphs in sections 3.x. Daily trending
+data used is provided in sections 4.x.
VPP worker on 1t1c
------------------