Products Cloud Switch
The Cubro Cloud Switch (CCS) is currently the end in this evolution from L4 NPB to an active network device with visibility functions included.
The CCS provides switching functions in layer 2 to 7 and at the same time visibility. This is possible because the packet forwarding is done in hardware. The switch infrastructure knows where the microservice is running, and can copy the relevant traffic before sending it over to the switch infrastructure - the probing system (virtual/real).
Even when the microservices are moved or scaled out the Cubro Solution can follow the application using the application ID across the data center.
- INBAND DYNAMIC VISIBILITY SERVICE STEERING
- CLOUD CENTRIC VISIBILITY
- APPLICATION BREATHING SUPPORT
- VISIBILITY IN HARDWARE (NO PERFORMANCE DEGRADATION IN THE SERVER)
- VISIBILITY BASED ON APPLICATION ID (MICRO SERVICE) LEVEL
- NO SOFTWARE CHANGES ON THE SERVER
- FULLY SOFTWARE AGNOSTIC
In a large network environment, there can be thousands of microservices to monitor with each service running on a separate server/rack/data center because of redundancy reasons. In the worst case, the “application” could be running all over the world in several data centers. There are even more difficulties because the applications can jump dynamically through the data centers. As a result, classical tapping is not possible anymore!
When the number of servers increases to thousands or tens of thousands and scale-outs and scale-ins happen regularly and across a large network it is a catch-up situation with SDV and there may be delays that are no more acceptable.
The only solution that can handle tens of thousands of servers with high cloud elasticity is a cloud switch that has tapping and network packet broker inbuilt. Cubro cloud switch (CCS) has all the necessary information about VM and container movements, scale-ins and scale-outs between hosts and is able to provide needed packets to the monitoring system.
The most advanced big data centers use the Cubro Cloud Switch (CCS), because it combines an advanced switching fabric with a visibility fabric.
Visibility must be a part of the network infrastructure in the cloud. For example in a data center with 10,000 servers or more physical tapping becomes expensive and virtual tapping unmanageable.
Cubro visibility solution follows the Cloud elasticity in real-time. This is only possible, if the visibility solution is a part of the network infrastructure, and gets Correlated Service Metadata out of the fabric.
In the Cubro solution each switch acts as a visibility node:
- High capability for load sharing
- The network itself can provide transport functions in the form of several types of tunnels forwarding the traffic to the monitoring tools.