WHY-OVS.rst

   1 ..
   2       Licensed under the Apache License, Version 2.0 (the "License"); you may
   3       not use this file except in compliance with the License. You may obtain
   4       a copy of the License at
   5
   6           http://www.apache.org/licenses/LICENSE-2.0
   7
   8       Unless required by applicable law or agreed to in writing, software
   9       distributed under the License is distributed on an "AS IS" BASIS, WITHOUT
  10       WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the
  11       License for the specific language governing permissions and limitations
  12       under the License.
  13
  14       Convention for heading levels in Open vSwitch documentation:
  15
  16       =======  Heading 0 (reserved for the title in a document)
  17       -------  Heading 1
  18       ~~~~~~~  Heading 2
  19       +++++++  Heading 3
  20       '''''''  Heading 4
  21
  22       Avoid deeper levels because they do not render well.
  23
  24 =================
  25 Why Open vSwitch?
  26 =================
  27
  28 Hypervisors need the ability to bridge traffic between VMs and with the outside
  29 world. On Linux-based hypervisors, this used to mean using the built-in L2
  30 switch (the Linux bridge), which is fast and reliable. So, it is reasonable to
  31 ask why Open vSwitch is used.
  32
  33 The answer is that Open vSwitch is targeted at multi-server virtualization
  34 deployments, a landscape for which the previous stack is not well suited. These
  35 environments are often characterized by highly dynamic end-points, the
  36 maintenance of logical abstractions, and (sometimes) integration with or
  37 offloading to special purpose switching hardware.
  38
  39 The following characteristics and design considerations help Open vSwitch cope
  40 with the above requirements.
  41
  42 The mobility of state
  43 ---------------------
  44
  45 All network state associated with a network entity (say a virtual machine)
  46 should be easily identifiable and migratable between different hosts. This may
  47 include traditional "soft state" (such as an entry in an L2 learning table), L3
  48 forwarding state, policy routing state, ACLs, QoS policy, monitoring
  49 configuration (e.g. NetFlow, IPFIX, sFlow), etc.
  50
  51 Open vSwitch has support for both configuring and migrating both slow
  52 (configuration) and fast network state between instances. For example, if a VM
  53 migrates between end-hosts, it is possible to not only migrate associated
  54 configuration (SPAN rules, ACLs, QoS) but any live network state (including,
  55 for example, existing state which may be difficult to reconstruct). Further,
  56 Open vSwitch state is typed and backed by a real data-model allowing for the
  57 development of structured automation systems.
  58
  59 Responding to network dynamics
  60 ------------------------------
  61
  62 Virtual environments are often characterized by high-rates of change. VMs
  63 coming and going, VMs moving backwards and forwards in time, changes to the
  64 logical network environments, and so forth.
  65
  66 Open vSwitch supports a number of features that allow a network control system
  67 to respond and adapt as the environment changes. This includes simple
  68 accounting and visibility support such as NetFlow, IPFIX, and sFlow. But
  69 perhaps more useful, Open vSwitch supports a network state database (OVSDB)
  70 that supports remote triggers. Therefore, a piece of orchestration software can
  71 "watch" various aspects of the network and respond if/when they change. This is
  72 used heavily today, for example, to respond to and track VM migrations.
  73
  74 Open vSwitch also supports OpenFlow as a method of exporting remote access to
  75 control traffic. There are a number of uses for this including global network
  76 discovery through inspection of discovery or link-state traffic (e.g. LLDP,
  77 CDP, OSPF, etc.).
  78
  79 Maintenance of logical tags
  80 ----------------------------
  81
  82 Distributed virtual switches (such as VMware vDS and Cisco's Nexus 1000V) often
  83 maintain logical context within the network through appending or manipulating
  84 tags in network packets. This can be used to uniquely identify a VM (in a
  85 manner resistant to hardware spoofing), or to hold some other context that is
  86 only relevant in the logical domain. Much of the problem of building a
  87 distributed virtual switch is to efficiently and correctly manage these tags.
  88
  89 Open vSwitch includes multiple methods for specifying and maintaining tagging
  90 rules, all of which are accessible to a remote process for orchestration.
  91 Further, in many cases these tagging rules are stored in an optimized form so
  92 they don't have to be coupled with a heavyweight network device. This allows,
  93 for example, thousands of tagging or address remapping rules to be configured,
  94 changed, and migrated.
  95
  96 In a similar vein, Open vSwitch supports a GRE implementation that can handle
  97 thousands of simultaneous GRE tunnels and supports remote configuration for
  98 tunnel creation, configuration, and tear-down. This, for example, can be used
  99 to connect private VM networks in different data centers.
 100
 101 Hardware integration
 102 --------------------
 103
 104 Open vSwitch's forwarding path (the in-kernel datapath) is designed to be
 105 amenable to "offloading" packet processing to hardware chipsets, whether housed
 106 in a classic hardware switch chassis or in an end-host NIC. This allows for the
 107 Open vSwitch control path to be able to both control a pure software
 108 implementation or a hardware switch.
 109
 110 There are many ongoing efforts to port Open vSwitch to hardware chipsets. These
 111 include multiple merchant silicon chipsets (Broadcom and Marvell), as well as a
 112 number of vendor-specific platforms. (The PORTING file discusses how one would
 113 go about making such a port.)
 114
 115 The advantage of hardware integration is not only performance within
 116 virtualized environments. If physical switches also expose the Open vSwitch
 117 control abstractions, both bare-metal and virtualized hosting environments can
 118 be managed using the same mechanism for automated network control.
 119
 120 Summary
 121 -------
 122
 123 In many ways, Open vSwitch targets a different point in the design space than
 124 previous hypervisor networking stacks, focusing on the need for automated and
 125 dynamic network control in large-scale Linux-based virtualization environments.
 126
 127 The goal with Open vSwitch is to keep the in-kernel code as small as possible
 128 (as is necessary for performance) and to re-use existing subsystems when
 129 applicable (for example Open vSwitch uses the existing QoS stack). As of Linux
 130 3.3, Open vSwitch is included as a part of the kernel and packaging for the
 131 userspace utilities are available on most popular distributions.