- play_arrow Working With Network Director
- play_arrow About Network Director
- play_arrow Installing Network Director
- play_arrow Accessing Network Director
- play_arrow Understanding Network Director System Administration and Preferences
- play_arrow Getting Started with Network Director
-
- play_arrow Working with the Dashboard
- play_arrow About the Dashboard
- play_arrow Using the Dashboard
- play_arrow Dashboard Widget Reference
-
- play_arrow Working in Deploy Mode
- play_arrow About Deploy Mode
- play_arrow Deploying and Managing Device Configurations
- Deploying Configuration to Devices
- Managing Configuration Deployment Jobs
- Deploy Configuration Window
- Importing Configuration Data from Junos OS Configuration Groups
- Enabling High-Frequency Traffic Statistics Monitoring on Devices
- Configuring Network Traffic Analysis
- Approving Change Requests
- Enabling SNMP Categories and Setting Trap Destinations
- Understanding Resynchronization of Device Configuration
- Resynchronizing Device Configuration
- Managing Device Configuration Files
- Creating and Managing Baseline of Device Configuration Files
- play_arrow Deploying and Managing Software Images
- play_arrow Managing Devices
- play_arrow Setting Up Zero Touch Provisioning for Devices
-
- play_arrow Monitoring Devices and Traffic
- play_arrow About Monitor Mode
- play_arrow Monitoring Traffic
- play_arrow Monitoring Client Sessions
- play_arrow Monitoring Devices
- play_arrow Monitoring and Analyzing Fabrics
- play_arrow Monitoring Virtual Networks
- play_arrow General Monitoring
- play_arrow Monitor Reference
- 802.11 Packet Errors Monitor
- Access vs. Uplink Port Utilization Trend Monitor
- Current Sessions Monitor
- Current Sessions by Type Monitor
- Error Trend Monitor
- Equipment Summary By Type Monitor
- Node Device Summary Monitor
- Port Status Monitor
- Port Status for IP Fabric Monitor
- Port Utilization Monitor
- Power Supply and Fan Status Monitor
- Resource Utilization Monitor for Switches, Routers, and Virtual Chassis
- Status Monitor for Junos Fusion Systems
- Status Monitor for Layer 3 Fabrics
- Status Monitor for Switches and Routers
- Status Monitor for Virtual Chassis
- Status Monitor for Virtual Chassis Members
- Top Talker - Wired Devices Monitor
- Traffic Trend Monitor
- Unicast vs Broadcast/Multicast Monitor
- Unicast vs Broadcast/Multicast Trend Monitor
- User Session Details Window
- Virtual Chassis Topology Monitor
- VC Equipment Summary By Type Monitor
-
- play_arrow Using Fault Mode
- play_arrow About Fault Mode
- play_arrow Using Fault Mode
- play_arrow Fault Reference
-
- play_arrow Working in Report Mode
- play_arrow About Report Mode
- play_arrow Creating and Managing Reports
- play_arrow Report Reference
-
- play_arrow Working with Network Director Mobile
- play_arrow About Network Director Mobile
- play_arrow Getting Started with Network Director Mobile
- play_arrow Working in the Network Director Mobile Dashboard Mode
- play_arrow Working in the Network Director Mobile Devices Mode
-
Understanding Class of Service (CoS) Profiles
When a network experiences congestion and delay, some packets must be prioritized to avoid random loss of data. Class of service (CoS) (also known as QoS) accomplishes this prioritization by dividing similar types of traffic, such as e-mail, streaming video, voice, large document file transfer, into classes. You then apply different levels of priority, such as those for throughput and packet loss, to each group, and thereby control traffic behavior. For example, when packets must be dropped, you can ensure that packet loss takes place according to your configured rules. CoS also enables you to rewrite the Differentiated Services code point (DSCP), IP precedence, or 802.1p CoS bits of packets exiting a specific interface, thus enabling you to tailor outgoing packets to meet the network requirements of remote peers.
How Would I Use CoS (also known as QoS)?
On an Ethernet trunk, you can mark frames with a class-of-service (CoS) value. CoS is used to define trunk connections as full-duplex, incoming only, or outgoing only.
Network devices such as routers and switches can be configured to use existing CoS values on incoming packets from other devices (trust mode), or can rewrite the CoS values to something completely different. Layer 2 markings also can extend to the WAN; for example, with a frame relay network. CoS is usually limited to use within an organization's intranet.
With legacy telephone systems, CoS can be used to define the permissions an extension will have on a private branch exchange (PBX) or Centrex. Some users might need extended voicemail message retention or the ability to forward calls to a cell phone, while others have no need to make calls outside the office. Permissions for a group of extensions can be changed by modifying a CoS variable applied to the entire group.
CoS configurations can be complicated, so unless it is required, we recommend that you do not alter the default class names or queue number associations.
How Do I Create CoS Groups?
Use 802.1Q tagged VLANs to group users and enable CoS to set priorities supported by downstream devices.
How Is CoS Different From QoS?
CoS operates only on 802.1Q VLAN Ethernet at the data link layer (layer 2), while quality-of-service (QoS) mechanisms operate at the IP network layer (layer 3). 802.1p Layer 2 tagging can be used by QoS to differentiate and shape network traffic.
What CoS Parameters Can I Control?
You can use CoS profiles to group a set of class of service (CoS) parameters and apply it to one or more interfaces. You can configure the following parameters within a CoS profile:
Classifiers—Packet classification refers to the examination of an incoming packet. This function associates the packet with a particular CoS servicing level.
Scheduler maps—Schedulers define the properties of output queues. These properties include the amount of interface bandwidth assigned to the queue, the size of the memory buffer allocated for storing packets, the priority of the queue, and the drop profiles associated with the queue. You associate the schedulers with forwarding classes by means of scheduler maps. You can then associate each scheduler map with an interface, thereby configuring the queues, packet schedulers, and tail drop processes that operate according to this mapping.
Rewrite values—A rewrite rule modifies the appropriate CoS bits in an outgoing packet. Modification of CoS bits enables the next downstream device to classify the packet into the appropriate service group. Rewriting or marking outbound packets is useful when the device is at the border of a network and must alter the CoS values to meet the policies of the targeted peer.
Traffic-control profile—Traffic-control profiles enable traffic limitation of a certain class to a specified bandwidth and burst size. Packets exceeding the limits can be discarded, or can be assigned to a different forwarding class, a different loss priority, or both.
What Are the Default CoS Traffic Types?
On EX Series switches, the system provides you with these four predefined traffic types—Data, Voice, Video, and Network Control—with these default traffic configuration and shaping details:
Data—Forwarding queue 0 (nd_best-effort), Buffer size 50%, Bandwidth reserved 30%
Voice—Forwarding queue 5 (nd_expedited-forwarding), Buffer size 20%, Bandwidth reserved 0%
Video—Forwarding queue 4 (nd_video-forwarding), Buffer size 20%, Bandwidth reserved 70%
Network Control—Forwarding queue 7 (nd_network-control), Buffer size 10%, Bandwidth reserved 0%
For Campus Switching ELS, the system provides you with these four predefined traffic types—Data, Voice, Video, and Network Control—with these default traffic configuration and shaping details:
Data—Forwarding queue 0 (nd_best-effort), Buffer size 50%, Bandwidth reserved 30%
Voice—Forwarding queue 1 (nd_expedited-forwarding), Buffer size 20%, Bandwidth reserved 0%
Video—Forwarding queue 2 (nd_video-forwarding), Buffer size 20%, Bandwidth reserved 70%
Network Control—Forwarding queue 3 (nd_network-control), Buffer size 10%, Bandwidth reserved 0%
For Campus Switching ELS with Hierarchal Port Scheduling (Juniper Networks EX2300 Ethernet switches), Network Director provides you with predefined forwarding classes—nd_cs_best-effort, nd_cs_video-forwarding, nd_cs_expedited-forwarding, and nd_cs_network-control. These forwarding classes are grouped under two priority groups—data_video_pg and voice_control_pg.
For Campus Switching ELS with Hierarchal Port Scheduling , you can modify and customize each of these priority groups and forwarding classes. For more details, see Creating and Managing Wired CoS Profiles.
Data Center Switching CoS Configuration
For data center switching devices, these additional CoS features are available:
Hierarchical Port Scheduling (ETS)—Hierarchical port scheduling (Enhanced Transmission Selection, or ETS) is a two-tier process that provides better port bandwidth utilization and greater flexibility to allocate resources to queues and to groups of queues.
Priority-based flow control (PFC)—A link-level flow control mechanism.
How Do I Implement Class of Service?
CoS can be implemented from the MSS CLI, from Network Director. RingMaster configures unicast traffic but does not configure multicast traffic. For directions to implement CoS from Network Director, see Creating and Managing Wired CoS Profiles.
Editing Discovered CoS Profiles
Duplicate scheduler configuration is deployed to the device when you edit a CoS profile that are automatically created by Network Director as part of device discovery or out-of-band changes. In CoS configuration, a single classifier can be associated to multiple ports regardless of the other CoS configuration. When Network Director discovers a device with such configuration it will create multiple profiles, based on the difference in other CoS configurations, and mapped to same classifier configuration. If you modify classifier settings in such a CoS profile that is created automatically by Network Director, Network Director cannot modify the configuration because it is mapped to multiple profiles. Whenever you modify such a CoS profile that is created automatically, Network Director will create new classifier settings configuration on the device and map the same to it, without affecting the existing classifier settings. Newly created classifier settings will have a name generated based on the profile name. Even if only one profile is mapped to the classifier settings, Network Director creates new classifier settings and the old settings are orphaned.
This behavior is applicable to both hierarchical and non hierarchical profiles, and is applicable for congestion notification profile name, traffic control profile name, scheduler map name, classifier name and rewrite rule settings.