VPP: Add comprehensive VPP documentation

This commit introduces extensive documentation for VyOS VPP (Vector Packet Processing)
Dataplane interfaces and features, including examples.

Features documented:
* VPP integration description, base requirements and limitations
* VPP dataplane core settings
* Features enabled in the dataplane: ACL, IPsec, NAT, sFlow
* VPP interface types: bonding, bridge, GRE, IPIP, kernel, loopback, VXLAN, XConnect

Note: This is an initial documentation version that will require ongoing improvements
based on user feedback and questions, real-world deployment testing and validation, additional
use cases and configuration scenarios discovered in production. Therefore, all pages are
marked as "need improvement".

1 files changed, 43 insertions, 0 deletions

diff --git a/docs/vpp/configuration/sflow.rst b/docs/vpp/configuration/sflow.rst
new file mode 100644
index 00000000..c28e517b
--- /dev/null
+++ b/docs/vpp/configuration/sflow.rst
@@ -0,0 +1,43 @@
+:lastproofread: 2025-09-04
+
+.. _vpp_config_sflow:
+
+.. include:: /_include/need_improvement.txt
+
+#######################
+VPP sFlow Configuration
+#######################
+
+VPP Dataplane in VyOS support sFlow for traffic monitoring and analysis.
+
+The VPP Dataplane integration works hand-in-hand with normal kernel sFlow agent, which is responsible for collecting and exporting sFlow samples. VPP itself is responsible for generating the samples.
+
+To enable sFlow in VPP, you first need to configure the service using the same steps as for normal kernel sFlow agent, as described in :doc:`/configuration/system/sflow`. Then you can enable sFlow on VPP interfaces.
+
+Then, you need to enable sFlow on the VPP interfaces you want to monitor. This is done using the following commands:
+
+.. cfgcmd::
+
+   set vpp sflow interface <interface-name>
+
+This will enable sFlow on the specified interface. You can repeat this command for each interface you want to monitor.
+
+.. note::
+
+   sFlow collects statistics only for traffic *received* on the interface. If you want to monitor traffic *sent* on the interface, you need to enable sFlow on the corresponding interface in the opposite direction.
+
+Optionally, you can specify the sampling rate for the interface using the following command:
+
+.. cfgcmd::
+
+   set vpp sflow sample-rate <rate>
+
+This will set the sampling rate for the specified interface. The default sampling rate is 1, which means that every packet is sampled. A higher sampling rate means that fewer packets are sampled, which can reduce the amount of data sent to the sFlow collector. This can be useful in high-traffic environments to reduce the load on the collector.
+
+Finally, you need to enable integration between VPP and the kernel sFlow agent using the following command:
+
+.. cfgcmd::
+
+   set system sflow vpp
+
+After this, collecting and exporting sFlow samples will be handled by the kernel sFlow agent, while VPP will generate the samples.