aboutsummaryrefslogtreecommitdiffstats
path: root/tests/performance/README.md
diff options
context:
space:
mode:
authorThomas Duval <thomas.duval@orange.com>2020-06-03 10:06:52 +0200
committerThomas Duval <thomas.duval@orange.com>2020-06-03 10:06:52 +0200
commit7bb53c64da2dcf88894bfd31503accdd81498f3d (patch)
tree4310e12366818af27947b5e2c80cb162da93a4b5 /tests/performance/README.md
parentcbea4e360e9bfaa9698cf7c61c83c96a1ba89b8c (diff)
Update to new version 5.4HEADstable/jermamaster
Signed-off-by: Thomas Duval <thomas.duval@orange.com> Change-Id: Idcd868133d75928a1ffd74d749ce98503e0555ea
Diffstat (limited to 'tests/performance/README.md')
-rw-r--r--tests/performance/README.md80
1 files changed, 0 insertions, 80 deletions
diff --git a/tests/performance/README.md b/tests/performance/README.md
deleted file mode 100644
index fcb80589..00000000
--- a/tests/performance/README.md
+++ /dev/null
@@ -1,80 +0,0 @@
-# Moon Yardstick/Bottlenecks Performance Tests
-
-The main objective of this document is to describe the performance tests for the Moon project/module.
-Moon is a security management platform which provides a set of security functions to project the underlying OPNFV infrastructure and/or VNFs.
-It is consisted of 2 parts: a master and a set of slaves. The master holds all security-related information and each slave only fetches and holds
-related information for its local usage from master.
-
-## Master Performance Tests
-### Pre-requisite
-- setup a Moon master service on a physical server
-- create a project in OpenStack/Keystone
-- create a MSL PDP with a model of 4 subject security levels and 4 object security levels, the MLS policy will be defined later
-
-### Policy Size Test
-Increase the number of users and resources N to find the limit of the security policy
-- create N users and N resources (VMs in our case) in this MLS security policy
-- sends 5 authz requests/second
-- gather performance metrics like CPU, memory, network usages
-Through the iteration, determine the maximal number of N to support 5 requests/second
-
-### PDP Number Test
-- setup 20 user and 20 resources (VMs in our case) for each MLS PDP
-- sends 5 authz requests/second for each MLS PDP
-- increase the number of PDP to test the maximal number of PDP on the master
-
-### Policy Size Test for 5 PDPs
-- setup 5 PDPs of N users and N resources (VMs in our case)
-- sends 5 authz requests/second for each MLS PDP
-- gather performance metrics like CPU, memory, network usages
-Through the iteration, determine the maximal user/resource number of these 5 PDPs
-
-### Policy Size Test for 10 PDPs
-- setup 10 PDPs of N users and N resources (VMs in our case)
-- sends 5 authz requests/second for each MLS PDP
-- gather performance metrics like CPU, memory, network usages
-Through the iteration, determine the maximal user/resource number of these 10 PDPs
-
-### Policy Size Test for 20 PDPs
-- setup 20 PDPs of N users and N resources (VMs in our case)
-- sends 5 authz requests/second for each MLS PDP
-- gather performance metrics like CPU, memory, network usages
-Through the iteration, determine the maximal user/resource number of these 20 PDPs
-
-
-## Master-Slave Performance Tests
-### Pre-requisite
-- setup a Moon master on a physical server
-- setup a Moon slave on a physical server
-- create a project in OpenStack/Keystone
-- create a MSL PDP with a model of 4 subject security levels and 4 object security levels, the MLS policy will be defined later on the master
-
-### Slave Policy Size Test
-Increase the number of users and resources N to find the limit of the security policy
-- create N users and N resources (VMs in our case) in this MLS security policy on the master
-- sends 5 authz requests/second to the slave
-- gather performance metrics like CPU, memory, network usages of the slave
-Through the iteration, determine the maximal number of N to support 5 requests/second of the slave
-
-### Slave PDP Number Test
-- setup 20 user and 20 resources (VMs in our case) for each MLS PDP on the master
-- sends 5 authz requests/second for each MLS PDP to the slave
-Through the iteration, determine the maximal number of PDP to support 5 requests/second of the slave
-
-### Slave Policy Size Test for 5 PDPs
-- setup 5 PDPs of N users and N resources (VMs in our case) on the master
-- sends 5 authz requests/second for each MLS PDP to the slave
-- gather performance metrics like CPU, memory, network usages of the slave
-Through the iteration, determine the maximal user/resource number of these 5 PDPs
-
-### Slave Policy Size Test for 10 PDPs
-- setup 10 PDPs of N users and N resources (VMs in our case) on the master
-- sends 5 authz requests/second for each MLS PDP to the slave
-- gather performance metrics like CPU, memory, network usages of the slave
-Through the iteration, determine the maximal user/resource number of these 10 PDPs
-
-### Slave Policy Size Test for 20 PDPs
-- setup 20 PDPs of N users and N resources (VMs in our case) on the master
-- sends 5 authz requests/second for each MLS PDP to the slave
-- gather performance metrics like CPU, memory, network usages of the slave
-Through the iteration, determine the maximal user/resource number of these 20 PDPs