Age | Commit message (Collapse) | Author | Files | Lines |
|
JIRA: YARDSTICK-1101
The traffic generated by the generator in bng-4 result in
non-symmetrical use of the cores on the SUT, as a whole
range of IP addresses is skipped.
Update cpe0 and cpe1 in gen files.
In addition, for the routing table (ipv4.lua) for the BNG
should be slighly different: while the BNG is a 4 ports
test, only 2 of the ports are used as outgoing routing
ports (the inet ports). Hence, the routing table should
only use the mac of those two ports. The bad configuration
(using 4 ports) result in an issue when BNG SUT and GEN
are connected through a switch, packets are sent by the
BNG SUT through the INET interfaces, but sometimes with
the MAC of the CPE interface...
Hence, packets are received on the wrong interface on the
Gen site (resulting in some interfaces, the CPE ones,
being overloaded)
Created ipv4_bng.lua and remove mac 0 and 2
Change-Id: I97019f0415d2f58d50ec855d04800eb705f41541
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
Added pktgen VNF topology, VNF descriptor and traffic profile files.
JIRA: YARDSTICK-1346
Change-Id: Ifb1824edb2dc8db1afa83a53573fb2ed1c2a79ea
Signed-off-by: Rodolfo Alonso Hernandez <rodolfo.alonso.hernandez@intel.com>
|
|
The acl_rules.yaml is unused, so renamed it to sample config.
Updated the config with example of possible ACL configuration.
JIRA: YARDSTICK-1304
Change-Id: If0537cf55f4c96f57844caed5bf1c9050647e0d5
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
Now "NSPerf" scenario will inform about the traffic generator PIDs after
setup process. With this information, IterationIPC runner will be able
to receive the messages sent by those traffic generators and control
the main iteration loop.
The following example, using vFW as VNF and OpenStack as context, makes
use of this new runner implementation:
/samples/vnf_samples/nsut/vfw/
tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex_iterationipc.yaml
JIRA: YARDSTICK-1127
Change-Id: I46b1368bc209680b88ff9fb5c3b9beadf6271ac9
Signed-off-by: Rodolfo Alonso Hernandez <rodolfo.alonso.hernandez@intel.com>
|
|
JIRA: YARDSTICK-1101
Problem was incorrect generator and changed the traffic profile to allow
a tolerated loss of 5%.
ARP packets are not managed (just passed thru the system) therefor the
tolerated loss must be higehr than normal.
Change-Id: I03bd62b397a66b50961e810d033c6894d0b3e9a6
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
JIRA: YARDSTICK-1105
Running a vPE test ends up with up to 25% drooped packets.
This is due to tolerated loss being incorrectly configured,
test duration was too short and test interval was too
short.
Note This is due to the fact that vPE use case have by
default a tolerated loss of 100%, Now changed to 0.001,
same as L2FWD
Also: Grafan dashboard has been updated to show all SUT
CPUS Utilization in 1 panel. And LINE rate MIN, MAX & TEST
Rate have been added
Change-Id: I7adae2199b3f656fe460705b6aeb3aa69c767d50
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
|
|
JIRA: YARDSTICK-1221
In order to increase accuracy of result the following are done :-
- Improve Measurement Accuracy (YARDSTICK-1212)
- Improve Sampling Interval (YARDSTICK-1219)
- Allow 4 PROX ports to be read simultaneously (YARDSTICK-1220)
This change does the following :-
- Stores LINE Rate statistics of Sample
- Requires a confirmation retry before deciding to increase or
decrease this is configurable.
- Allows the user to disable Sampling or specify a sample interval
- Added Code Coverage of ProxDurationRunner based on YARDSTICK-1199
Change-Id: I27242ac1849c9a2712866385b5fbc05977c71516
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
The uplink and downlink interfaces for the 4 port
test cases were inconsistent and fixed in this commit.
This affects all the 4 port PROX test cases.
Tested on: Heat L2fwd-4 and L3fwd-4 test case.
JIRA: YARDSTICK-1185
Change-Id: Ia2ce7ed0dc395b812f84ede94e259b42dc833579
Signed-off-by:Abhijit Sinha <abhijit.sinha@intel.com>
|
|
|
|
This patch allows user to configure ACL/vFW SampleVNF ACL
via configuration file provided in TC definition. The
Yardstick applies the rules to SampleVNF from specified config
file + rules generated by Yardstick (default rules).
The example of SampleVNF ACL CLI commands generated/applied
by Yardstick can be found at (using default ACL config file):
ACL VNF ACL CLI commands:
acl/tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex.yaml
http://paste.openstack.org/show/723303/
vFW VNF ACL CLI commands:
vfw/tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex.yaml
http://paste.openstack.org/show/723304/
Change-Id: I76a630261a982083b628e3985fc3bec14ca495db
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
|
|
Added default values for pmd/lcore mask to the SampleVNF test case for
Trex TG.
lcore mask - specifies the CPU cores used for non-datapath OVS-DPDK
threads.
pmd mask - specifies the CPU cores used for data path packet processing.
Updated documentation for Standalone OVS-DPDK context.
JIRA: YARDSTICK-1246
Change-Id: I8d008a185c94d2047ffa4769d7974bc4d283fd7f
Signed-off-by: Stepan Andrushko <stepanx.andrushko@intel.com>
|
|
JIRA: YARDSTICK-1211
Change-Id: I0899d470ca5a0ec7d42d36a9ff9c39cc7369310c
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
|
|
Add PROX and vFW scale out TCs with availability zone
configuration. To be able to specify the availability
zone for specific VNF or TG the following example of
command line option shoud be used:
yardstick -d task start --task-args='{
"num_vnfs": 2, "availability_zone":
{
"vnf_0": "az_0", "tg_0": "az_0",
"vnf_1": "az_1", "tg_1": "az_1",
... }}' <SCALE-OUT-TC-FILE>
The steps to run new scale-out TCs is documented in the
userguide.
JIRA: YARDSTICK-1203
Change-Id: Ib3878485c5483981fe9b48ab02995712bcc31ae5
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
The Prox L2-fwd test case has a bad packet, which fails
the test case to run in OpenStack context as it's dropped
by the neutron.
This fix is needed to successfully run the L2fwd 2 and
4 port test cases in heat based test cases.
Change-Id: I86df31a6df1eaa6f968e46af159ac6a0157f402c
JIRA: YARDSTICK-1164
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
JIRA: YARDSTICK-1053
Those three yaml files are for bottlenecks support offiline flavor
The ssh time out problem will fix in changing the duration time
Change-Id: Ifd317404a820d0446ffab5070211e48def1a0d2f
Signed-off-by: liyin <liyin11@huawei.com>
|
|
|
|
JIRA: YARDSTICK-1037
NSB PROX L3FWD was incorrect. The routing table (ipv4.lua) was incorrect.
Also a specific ipv4 for 2 port configuration is required.
Change-Id: I3a5fd1ea1d1ddf74cbdd6a5d9e7ced42d3c1d6e9
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
JIRA: YARDSTICK-1035
Do not hardcode NIC/interface speed in PROX test. Test assumes
NIC used is 10Gbps. This is incorrect. It could support 1Gbps,
10Gbps, 25Gbps, 40Gbps or something else.
This is used to calculate pps (Packets Per Second)
In Baremetal the NIC speed could be extracted. however when
run on a virtual machine this is not possible.
Solution:
Add in options section of test file.
eg.
Options:
interface_speed_gbps: 10
Where 10 refers to a 10Gbps. In a setup where multiple interfaces
are used. This will refer to the speed of the slowest connection.
Change-Id: I89ab16479a2cdd1d79e52cbcc5a972762c60d057
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
These scale-up TCs replace PROX static TCs which will be removed
by separate commit once those new TCs are merged.
Change-Id: I892a81c4bfd199782fb561faccc11e70a49675ae
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
Added topology and traffic profile templates
Added support for using JinJa2 templates in topology definition
Added support for static pipeline configs for SampleVNFs
JIRA: YARDSTICK-1043
Change-Id: Iab99fd5b5ad69ca32ee70b9fe47779387ad27e7f
Signed-off-by: Chornyi, TarasX <tarasx.chornyi@intel.com>
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
|
|
JIRA: YARDSTICK-1036
This stores a number of extra counters in influxdb for Prox test cases.
It also stores existing counters with a "succcess_" tag.
Previously throughput where stored without success or failure indication.
Also "Result_" counters are also stored.
These can now be used by Grafana to graph output.
Change-Id: Ie5636c14ecbab1b53a988bdfbd47ddd1fcdbd695
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
Updated vfw/tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex.yaml TC to be
able to configure segmentation_id from commmand line.
E.g.:
yardstick -d task start --task-args='{"provider": "sriov", \
"segmentation_id": 100}' ...
JIRA: YARDSTICK-1004
Change-Id: I53daae3468de2a1f523bd96817143a5f809238a2
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
|
|
The "context:file" of tc_prox_baremetal_lw_aftr
has an incorrect path, compaired to all other
prox test cases.
Change-Id: Id24e74905f36bc84fdf93d1a2966211ecad55c8b
Signed-off-by: Xavier Simonart <xavier.simonart@intel.com>
|
|
Updated vfw/tc_heat_rfc2544_ipv4_1rule_1flow_64B_trex.yaml TC to be
able to run it in heat context with SRIOV VF ports configured in
OpenStack.
E.g.:
yardstick task start --task-args='{"provider": "sriov"}' ...
yardstick task start --task-args='{"provider": "sriov"}\
physical_networks: ["physnet1", "physnet2"]}' ...
Console log: http://paste.openstack.org/show/666931/
Result output: http://paste.openstack.org/show/666930/
JIRA: YARDSTICK-1004
Change-Id: If0dabd0791e4e295782fee0f3de1536474218876
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
JIRA: YARDSTICK-1003
- There are redundant files for Prox vnf descriptors
tg and vnf for 1, 2 and 4 port set-up.
- We can use one vnf descriptor file per prox tg and prox vnf,
reducing the number of redundant files.
- Tested on Prox BM L2fwd 2 and 4 port test case.
Change-Id: I4c61fd11725121f19392443460ac2ad39c934e2d
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
second, Throughput and Transactions per second for various http locator image sizes including 1b, 4k, 64k, 256k, 1024k"
|
|
Add sample OpenStack/SRIOV RFC2544 based test case which can be
run using external Trex.
Console log: http://paste.openstack.org/show/658002/
Result output: http://paste.openstack.org/show/658004/
JIRA: YARDSTICK-967
Change-Id: Id0e2d6d47f459c7e6a3370735bab74f7ce502b24
Signed-off-by: Mytnyk, Volodymyr <volodymyrx.mytnyk@intel.com>
|
|
JIRA: YARDSTICK-965
Removed the redundant files and directory from
Prox test folder.
Change-Id: I6167f0a4ccdd581153dad3ec9f5eaaf2619aacbd
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
|
|
A generic throughput test case that can be used as a
stub code for a Linux-based VNF configured as an L3 forwarder.
Supported context:
* Standalone OVSDPDK and SRIOV
* Baremetal
Code changes:
* Allow pmd-cpu-mask and lcore mask for OVS DPDK
* router_vnf.py - configures interface IP addresses and static arp entries
using ip command
* NFVi KPIs
* Allow cputune tag for standalone context to be able to PIN on NUMA 1 cpus
SRIOV Test cases:
* RFC2544 Ethernet framesizes, 128K Flows
* 2,4 and 6 ports
* 2 and 3 vcpus per port
*
OVSDPDK Test cases:
* RFC2544 Ethernet framesizes, 128K Flows
* 2 and 4 ports
* 2 vcpus per port
* 2 PMD threads per port
TODO:
* Documentation
* Add 6 ports tests
References:
* router_vnf.py is based on sample_vnf.py
* tc_*.yaml files are based on acl/vfw test case files
Added unitests
Added get_stats to parse ip -s link
Change-Id: Id1b969d5420dfcab7c1e695acbd2cd1655747efe
Signed-off-by: Dino Simeon Madarang <dinox.madarang@intel.com>
Signed-off-by: Kavindya Deegala <kavindya.s.deegala@intel.com>
Reviewed-by: Alain Jebara <alain.jebara@intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Reviewed-by: Emma Foley <emma.l.foley@intel.com>
Reviewed-by: Rodolfo Alonso Hernandez <rodolfo.alonso.hernandez@intel.com>
Reviewed-by: Ross Brattain <ross.b.brattain@intel.com>
Reviewed-by: Edward MacGillivray <edward.s.macgillivray@intel.com>
Signed-off-by: Dhaval Patel <dhaval.r.patel@intel.com>
Signed-off-by: Ross Brattain <ross.b.brattain@intel.com>
|
|
Changed private to uplink and public to downlink
Change-Id: Id9273d9489f58bd45b60a00fc9e5bcbe9e136c84
Signed-off-by: Phani Kiran Thaticharla <phani.kiran.thaticharla@intel.com>
Reviewed-by: Oscar Medina-Duran <oscar.medina-duran@intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Reviewed-by: Ross Brattain <ross.b.brattain@intel.com>
|
|
second, Throughput and
Transactions per second for various http locator image sizes including 1b, 4k, 64k, 256k, 1024k
Change-Id: I253d1fa20a0324bc9efeaa4da1a24560e644734c
Signed-off-by: Phani Kiran Thaticharla <phani.kiran.thaticharla@intel.com>
Reviewed-by: Oscar Medina-Duran <oscar.medina-duran@intel.com>
Reviewed-by: Deepak S <deepak.s@linux.intel.com>
Reviewed-by: Ross Brattain <ross.b.brattain@intel.com>
|
|
|
|
|
|
YARDSTICK-839
In a scale-up test the VNF number can start from
0 to MAX (num_vnf-1)
The NSB PROX BM test nodes are currently marked as:
tg__0: trafficgen_1.yardstick
vnf__0: vnf.yardstick
They need to be changed to the below:
tg__0: tg_0.yardstick
vnf__0: vnf_0.yardstick
Changes done are:
- All BM Prox test cases have tg_0 and vnf_0 as default
test nodes
- The prox-baremetal-1/2/4.yaml sample pod file
- Cleaned Up some unused Prox test/config files
Change-Id: I819011bcb35eda62b17dd3e1035918918c582b13
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
Use Apache 2.0
Change-Id: I199d09f04a5c84a25bd91dceba5e4ed1f93075f6
Signed-off-by: Ross Brattain <ross.b.brattain@intel.com>
|
|
|
|
Changed Copyright notice on tuples.lua from "Intel Copyright" to Apache 2.0
Change-Id: I85e6f9eb86c116d552f86cddf05b3b368b32f776
Signed-off-by: Daniel Martin Buckley <daniel.m.buckley@intel.com>
|
|
JIRA: YARDSTICK-802
Addition of PROX LW_AFTR basked on PROX/DATS v037 test_104_lw_aftr.py
- This support BM and Openstack Heat
- This supports 4 Ports ONLY
- Grafana Dashboards included
- Code Coverage / Unit testing
Change-Id: If2170ab458bf687256d5f1a1e840a3b9d2788ef7
Signed-off-by: Daniel MArtin Buckley <daniel.m.buckley@intel.com>
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
- concurrency
- connections/second
- throughput
Change-Id: Idb977462952fba2e6a9684e916dfeca810f1ee94
Signed-off-by: Jennifer Li <jennifer.li@intel.com>
|
|
rename xe0, xe1 to uplink_0, downlink_0
enumerate tgs/vnfs starting from 0.
add disable_dhcp
Change-Id: Ia4f08b3ad5ab298192c249d04b10a6efbe4f210e
Signed-off-by: Ross Brattain <ross.b.brattain@intel.com>
|
|
As per defect:
- the "context:" stanza of every tc_prox_baremetal*.yaml"
file has an incorrect pathname for the config file
- tc_prox_baremetal_binsearch.yaml and
tc_prox_baremetal_ramp.yaml refer to
"prox-baremetal.yml", which idoes not exist
- incorrect prox bin path
Resolution:
- Unify the pod file name and path for all BM tests
- Prox bin path changed to "/opt/nsb_bin/prox"
Change-Id: I16b25189104ad5585ddc1c2c0dd71b3c4c2aecd3
Signed-off-by: Abhijit Sinha <abhijit.sinha@intel.com>
|
|
|
|
Change-Id: I6bd8a87f70954f288975e3e73e3cfbd36ae561dc
Signed-off-by: Ross Brattain <ross.b.brattain@intel.com>
|
|
|