Reference Guide

• A combination of the previous two conditions
One possible impact of this behavior change is that if the amount of traffic to a destination is higher than
the volume of traffic that can be carried over one path, a portion of that traffic might be dropped for a
short duration (30-60 seconds) after the system comes up.
Delayed Installation of ECMP Routes Into BGP
The current FIB component of Dell Networking OS has some inherent inefficiencies when handling a
large number of ECMP routes (i.e., routes with multiple equal-cost next hops). To circumvent this for the
configuration of fast boot, changes are made in BGP to delay the installation of ECMP routes. This is done
only if the system comes up through a fast boot reload. The BGP route selection algorithm only selects
one best path to each destination and delays installation of additional ECMP paths until a minimum of 30
seconds has elapsed from the time the first BGP peer is established. Once this time has elapsed, all routes
in the BGP RIB are processed for additional paths.
While the above change will ensure that at least one path to each destination gets into the FIB as quickly
as possible, it does prevent additional paths from being used even if they are available. This downside has
been deemed to be acceptable.
RDMA Over Converged Ethernet (RoCE) Overview
This functionality is supported on the S6000 platform.
Remote direct memory access (RDMA) reduces both CPU cycles and latency. RDMA over converged
Ethernet (RoCE) implements IB over Ethernet. RRoCE sends InfiniBand (IB) packets over IP. IB supports
input and output connectivity for the internet infrastructure. InfiniBand enables the expansion of network
topologies over large geographical boundaries and the creation of next-generation I/O interconnect
standards in servers. Although the endpoints or the destination servers generate such RRoCE packets,
from the switch's perspective, RRoCE is processed as an IP packet.
RRoCE packets are received and transmitted on specific interfaces called lite-subinterfaces. These
interfaces are similar to the normal Layer 3 physical interfaces except for the extra provisioning that they
offer to enable the VLAN ID for encapsulation.
You can configure a physical interface or a Layer 3 Port Channel interface as a lite subinterface. When
you configure a lite subinterface, only tagged IP packets with VLAN encapsulation are processed and
routed. All other data packets are discarded.
To provide lossless service for RRoCE, the QoS service policy must be configured in the ingress and
egress directions on lite subinterfaces.
A normal Layer 3 physical interface processes only untagged packets and makes routing decisions based
on the default Layer 3 VLAN ID (4095).
To enable routing of RRoCE packets, the VLAN ID is mapped to the default VLAN ID of 4095 using VLAN
translation. After the VLAN translation, the RRoCE packets are processed in the same way as normal IP
packets that a Layer 3 interface receives and routes in the egress direction. At the egress interface, the
VLAN ID is appended to the packet and transmitted out of the interface as a tagged packet with the
dot1Q value preserved.
340
Flex Hash and Optimized Boot-Up