Simplified, High-Performance 10GbE Networks Based on a Single Virtual Distributed Switch, Managed by VMware* vSphere 5.1

5.1 Network Resource Management
When network resource management is enabled, VDS trafc is
divided into the network resource pools shown in Table 1.
The VDS (but not the VSS) supports vSphere Network I/O
Control (NetIOC), which provides for the use of limits and shares
parameters to control the allocation of shared physical network
resources that are contended for by multiple trafc types on the
same network pipe. NetIOC is a powerful feature that can make
vSphere deployments even more suitable for I/O-consolidated
data centers. The following best practices can help optimize the
use of this feature:
• Best practice 1: For bandwidth allocation, use shares
instead of limits, since the former has greater flexibility
for unused capacity redistribution. Partitioning the available
network bandwidth among different types of network traffic
flows using limits has shortcomings. In other words, limits
impose hard limits on the amount of the bandwidth usage by a
traffic flow even when there is network bandwidth available.
• Best practice 2: Consider imposing limits on a given resource
pool when you are concerned about physical switch or physical
network capacity. Limiting the bandwidth usage of specific
resource pools at the VMware ESX* host level can help prevent
the possibility of jeopardizing performance for other flows going
through the same points of contention.
• Best practice 3: Set the corresponding resource-pool shares
to the predefined default shares value for VMware FT,
because fault tolerance is a latency-sensitive traffic flow.
• Best practice 4: Use Load-Based Teaming (LBT) as your VDS
teaming policy while using NetIOC, in order to maximize the
networking capacity utilization.
• Best practice 5: Use the DV Port Group and Traffic Shaper
features offered by the VDS to maximum effect when
configuring the VDS. Configure each of the traffic flow types
with a dedicated DV Port Group. Use DV Port Groups as a
means to apply configuration policies to different traffic flow
types, and more important, to provide additional Rx bandwidth
controls through the use of Traffic Shaper. For instance, you
might want to enable Traffic Shaper for the egress traffic on
the DV Port Group used for vMotion. This can help in situations
when multiple vMotion migrations initiated on different vSphere
hosts converge to the same destination vSphere server.
5.2 Network Path Redundancy
Providing network path redundancy between cluster nodes
is important, especially for vSphere HA reliability. A single
management network constitutes a single point of failure and
can result in failovers even though only the network has failed.
Possible failures include NIC failures, network cable failures,
network cable removal, and switch resets. Network architects
must consider these possible sources of failure between hosts and
try to minimize them, typically by providing network redundancy.
Network redundancy can be implemented at the NIC level with
NIC teaming or at the management network level. In most
implementations, NIC teaming provides sufcient redundancy;
management network redundancy is a valuable option in cases
where additional redundancy is desirable. Best practices call for
conguring the fewest possible hardware segments between
the servers in a cluster, to limit single points of failure and packet
delays caused by excessive numbers of hops.
• Network redundancy using NIC teaming. Using a team of
two NICs connected to separate physical switches improves
the reliability of a management network. Because servers
connected through two NICs (and through separate switches)
have two independent paths for sending and receiving
heartbeats, the cluster is more resilient. To configure a NIC team
for the management network, configure the vNICs in vSwitch
configuration for Active or Standby configuration.
Network I/O Control – Quality-of-Service Traffic Types
• VMware vMotion* traffic • Virtual machine traffic
• Management traffic • User-defined network resource pools
• NFS traffic • VMware vSphere* replication traffic
• iSCSI traffic • Fault-tolerant traffic
Note: The iSCSI traffic resource pool shares do not apply to
iSCSI traffic on a dependent hardware iSCSI adapter.
9
Simplied, High-Performance 10GbE Networks Based on a Single Virtual Distributed Switch, Managed by VMware vSphere* 5.1