diff options
author | Vratko Polak <vrpolak@cisco.com> | 2019-05-16 10:43:31 +0200 |
---|---|---|
committer | Tibor Frank <tifrank@cisco.com> | 2019-05-16 08:58:49 +0000 |
commit | 7c91ed6b45887fc8b31ce40595aab93352964eba (patch) | |
tree | 2ddb17a680a5124b88674a2ebfed69ca3570879a | |
parent | 09e63047565022fcf57d5b8ee208166e6e4c0d24 (diff) |
Improve soak graphs and content
+ Fix duration, mention number of tests and runs.
- Hover information still feels wrong.
+ Mention "critical rate" instead of "throughput" where possible.
+ Mention lower bounds are displayed.
+ Rename "plrsearch_algorithm" label to "PLRsearch".
+ Because the previous link did not work. Due to underscore?
+ Rename "SOAK" to "soak" or "Soak" (except when it is a tag).
Change-Id: I9c079f8c2e93d96249d4b60893469c9257d4f3db
Signed-off-by: Vratko Polak <vrpolak@cisco.com>
(cherry picked from commit 9a6b6ed77705147032abee3fc98a2f7d370427c0)
3 files changed, 23 insertions, 21 deletions
diff --git a/docs/report/introduction/methodology_data_plane_throughput/methodology_plrsearch.rst b/docs/report/introduction/methodology_data_plane_throughput/methodology_plrsearch.rst index 4aa1019b4c..88e6f901c8 100644 --- a/docs/report/introduction/methodology_data_plane_throughput/methodology_plrsearch.rst +++ b/docs/report/introduction/methodology_data_plane_throughput/methodology_plrsearch.rst @@ -1,4 +1,4 @@ -.. _plrsearch_algorithm: +.. _PLRsearch: PLRsearch ^^^^^^^^^ diff --git a/docs/report/vpp_performance_tests/soak_tests/index.rst b/docs/report/vpp_performance_tests/soak_tests/index.rst index bc236c0298..faa4675418 100644 --- a/docs/report/vpp_performance_tests/soak_tests/index.rst +++ b/docs/report/vpp_performance_tests/soak_tests/index.rst @@ -33,7 +33,9 @@ Soak Tests ========== -Long duration soak tests are executed using `plrsearch_algorithm`_. +Long duration (2 hours per test) soak tests are executed +using `PLRsearch`_ algorithm. As the test take long time, only 10 test +were executed, two runs each. Additional information about graph data: @@ -41,18 +43,18 @@ Additional information about graph data: #. **X-axis Labels**: indices of test suites. -#. **Y-axis Labels**: estimated intervals for packet throughput value in [Mpps]. +#. **Y-axis Labels**: estimated lower bounds for critical rate value in [Mpps]. #. **Graph Legend**: list of X-axis indices with CSIT test suites. -#. **Hover Information**: lists minimum, first quartile, median, +#. **Hover Information**: in general lists minimum, first quartile, median, third quartile, and maximum. If either type of outlier is present the whisker on the appropriate side is taken to 1.5×IQR from the quartile (the "inner fence") rather than the max or min, and individual outlying data points are displayed as unfilled circles (for suspected outliers) or filled circles (for outliers). (The "outer fence" is 3×IQR from the quartile.) - It also lists the lower bounds for the estimated throughput value. + When number of samples is low, some values are not displayed. .. note:: diff --git a/resources/tools/presentation/specification.yaml b/resources/tools/presentation/specification.yaml index 74e33c10fe..94ffec90f5 100644 --- a/resources/tools/presentation/specification.yaml +++ b/resources/tools/presentation/specification.yaml @@ -231,8 +231,8 @@ plot-vpp-soak-2n-skx: # TODO: Add 1904 data csit-vpp-perf-verify-1904-2n-skx: - - 70 # SOAK - - 71 # SOAK + - 70 # soak + - 71 # soak plot-vpp-nfv-2n-skx-ndrpdr-64b: csit-vpp-perf-verify-1904-2n-skx: @@ -610,8 +610,8 @@ - 32 # NDRPDR sel vpp-soak-vs-ndr-2n-skx-cmp: csit-vpp-perf-verify-1904-2n-skx: - - 70 # SOAK - - 71 # SOAK + - 70 # soak + - 71 # soak vpp-soak-vs-ndr-2n-skx: csit-vpp-perf-verify-1904-2n-skx: - 12 # NDRPDR sel @@ -624,8 +624,8 @@ - 30 # NDRPDR sel - 31 # NDRPDR sel - 32 # NDRPDR sel - - 70 # SOAK - - 71 # SOAK + - 70 # soak + - 71 # soak vpp-performance-changes-3n-hsw-mrr-h1: csit-vpp-perf-verify-1810-3n-hsw: @@ -957,7 +957,7 @@ size: 16 zeroline: False yaxis: - title: "<b>Packet Throughput [Mpps]</b>" + title: "<b>Critical Rate Lower Bound [Mpps]</b>" titlefont: size: 16 gridcolor: "rgb(230, 230, 230)" @@ -1523,8 +1523,8 @@ - 69 # NFV NDRPDR IMIX - 80 # NFV MRR NDRPDR 64b IMIX - 81 # NFV MRR NDRPDR 64b IMIX - - 70 # SOAK - - 71 # SOAK + - 70 # soak + - 71 # soak # csit-ligato-perf-verify-1901-3n-hsw: # - 10 # NDRPDR full @@ -1670,7 +1670,7 @@ ### T A B L E S ### ################################################################################ -# VPP SOAK vs NDR +# VPP Soak vs NDR - type: "table" title: "VPP Soak vs NDR 2n-skx" algorithm: "table_soak_vs_ndr" @@ -4570,9 +4570,9 @@ ### P L O T S ### ################################################################################ -# Soak test - 30min Soak Test (PLRsearch), boxes +# Soak Test (PLRsearch), boxes - type: "plot" - title: "VPP Throughput: 30min Soak Test (PLRsearch) boxes" + title: "VPP Critical rate: 2 Hours Soak Test (PLRsearch)" algorithm: "plot_performance_box" output-file-type: ".html" output-file: "{DIR[STATIC,VPP]}/soak-test-1" @@ -4594,12 +4594,12 @@ boxpoints: "all" whiskerwidth: 0 layout: - title: "30min Soak Test (PLRsearch)" + title: "2 Hours Soak Test (PLRsearch)" layout: "plot-soak-throughput" -# Soak test - 30min Soak Test (PLRsearch), boxes +# Soak Test (PLRsearch), boxes - type: "plot" - title: "VPP Throughput: 30min Soak Test (PLRsearch) boxes" + title: "VPP Critical rate: 2 Hours Soak Test (PLRsearch)" algorithm: "plot_performance_box" output-file-type: ".html" output-file: "{DIR[STATIC,VPP]}/soak-test-2" @@ -4619,7 +4619,7 @@ boxpoints: "all" whiskerwidth: 0 layout: - title: "30min Soak Test (PLRsearch)" + title: "2 Hours Soak Test (PLRsearch)" layout: "plot-soak-throughput" ################################################################################ |