summaryrefslogtreecommitdiffstats
path: root/docs
diff options
context:
space:
mode:
authorMofassirArif <Mofassir_Arif@dellteam.com>2016-01-25 03:56:21 -0800
committerMofassirArif <Mofassir_Arif@dellteam.com>2016-01-25 03:56:21 -0800
commit0e745438997721838b763b089b560ddb0c53d98b (patch)
treee6ff95f0f76b75c4bade766cd13ca86ac9b024db /docs
parent971a7c98515a9d83661f5e423f7e8390f35dca59 (diff)
update: address for testing db
Change-Id: I48b30fffd1de2512e4f46ee951a85299edc06a1c Signed-off-by: MofassirArif <Mofassir_Arif@dellteam.com>
Diffstat (limited to 'docs')
-rw-r--r--docs/network_testcases.rst31
1 files changed, 16 insertions, 15 deletions
diff --git a/docs/network_testcases.rst b/docs/network_testcases.rst
index ac68b11b..1c6fb910 100644
--- a/docs/network_testcases.rst
+++ b/docs/network_testcases.rst
@@ -7,11 +7,12 @@ Network Throughput for VMs
Network Throughput for Compute Nodes
For the throughout of the compute nodes we simply go into the systems-under-test
-and install iperf3 on the nodes. One of the SUTs is used a server and the other as a
-client. The client pushes traffic to the server for a duration specified by the user
-configuration file for iperf. These files can be found in the test_cases/{POD}/network/
-directory. The bandwidth is limited only by the physical link layer speed available to the server.
-The result file inlcudes the b/s bandwidth and the CPU usage for both the client and server.
+ and install iperf3 on the nodes. One of the SUTs is used a server and the
+ other as a client. The client pushes traffic to the server for a duration specified by
+ the user
+ configuration file for iperf. These files can be found in the test_cases/{POD}/network/
+ directory. The bandwidth is limited only by the physical link layer speed available to the server.
+ The result file inlcudes the b/s bandwidth and the CPU usage for both the client and server.
For the VMs we are running two topologies through the framework.
@@ -19,20 +20,20 @@ For the VMs we are running two topologies through the framework.
2: VMs on different compute nodes
QTIP framework sets up a stack with a private network, security groups, routers and attaches the VMs to this network. Iperf3 is installed
-on the VMs and one is assigned the role of client while other serves as a server. Traffic is pushed
-over the QTIP private network between the VMs. A closer look in needed to see how the traffic actually
-flows between the VMs in this configuration to understand what is happening to the packet as traverses
-the openstack network.
+ on the VMs and one is assigned the role of client while other serves as a server. Traffic is pushed
+ over the QTIP private network between the VMs. A closer look in needed to see how the traffic actually
+ flows between the VMs in this configuration to understand what is happening to the packet as traverses
+ the openstack network.
The packet originates from VM1 and its sent to the linux bridge via a tap interface where the security groups
-are written. Afterwards the packet is forwarded to the Integration bridge via a patch port. Since VM2 is also connected
-to the Integration bridge in a similar manner as VM1 so the packet gets forwarded to the linux bridge connecting
-VM2. After the linux bridge the packet is sent to VM2 and is recieved by the Iperf3 server. Since no physical link is
-involved in this topology, only the OVS (Integration bridge) is being benchmarked and we are seeing bandwidth in the range
-of 14-15 Gbps.
+ are written. Afterwards the packet is forwarded to the Integration bridge via a patch port. Since VM2 is also connected
+ to the Integration bridge in a similar manner as VM1 so the packet gets forwarded to the linux bridge connecting
+ VM2. After the linux bridge the packet is sent to VM2 and is recieved by the Iperf3 server. Since no physical link is
+ involved in this topology, only the OVS (Integration bridge) is being benchmarked and we are seeing bandwidth in the range
+ of 14-15 Gbps.
For the topology where the VMs are spawned on different compute nodes, the path the packet takes becomes more cumbersome.
-The packet leaves a VM and makes its way to the Integration Bridge as in the first topology however the integration bridge
+The packet leaves a VM and makes its way to the Integration Bridge as in the first topology however the integration bridge
forwards the packet to the physical link through the ethernet bridge. The packet then gets a VLAN/Tunnel depending on the network
and is forwarded to the particular Compute node where the second VM is spwaned. The packets enter the compute node through the physical
ethernet port and makes its way to the VM through the integration bridge and linux bridge. As seen here the path is much more involved