Shulou(Shulou.com)11/24 Report--

"one arrow is easy to break, but ten arrows are hard to break."

This is a proverb story often heard when we were young, which literally means that one arrow is easy to break, but multiple arrows tied together are difficult to destroy, which tells us the importance of "unity and cooperation".

In the field of load bearing, there is also the wonderful use of this story. Let's take a look at it together.

Single link, easy to break in the early stage of the development of IP network, two network devices (An and B) generally communicate by establishing a single link.

Obviously, using a single-link connection has the following drawbacks:

There is a bandwidth bottleneck, and the total bandwidth between devices An and B is equal to the bandwidth of this single link.

There is no redundant backup of the link, and if there is a problem with this link, the communication between devices An and B may be interrupted.

Even if the single link has obvious defects, but in that era of low requirements for the network, it can still meet the needs of the public.

With the development of network scale, single link can no longer meet the requirements of network bandwidth and reliability, so people think of bundling multiple links like arrows. As a result, link aggregation technology came into being.

Link aggregation, ten-arrow hard-to-break link aggregation forms an aggregation group by "bundling" multiple links between two devices, and multiple links in the aggregation group can be regarded as a logical link.

At this point, the traffic between the two devices can be shared on the links within the aggregation group.

So, what benefits can link aggregation bring to the network?

Increased network bandwidth bundles multiple links between An and B into a logical link, and the bundled bandwidth is the sum of the bandwidth of all links.

For example, if there are three links between An and B for link aggregation, and the bandwidth of each link is 10 Gbps, the maximum bandwidth of this aggregation group can reach 30 Gpbs.

Improve the reliability of the network connection if a link between An and B fails and is interrupted, the traffic will be automatically redistributed among the remaining links and will not cause the traffic interruption between An and B.

The realization of load-balanced link aggregation of traffic can evenly distribute the traffic between An and B among all member links, minimizing the risk of traffic blocking links in each member link.

Avoid the generation of two-layer loop when the links between An and B adopt link aggregation, these links no longer work independently and become a logical link externally. Therefore, there is no loop without using STP (Spanning Tree Protocol, spanning Tree Protocol), which effectively avoids the risk of two-layer loop between An and B.

Based on the above advantages, link aggregation has been widely used in IP networks.

MC-LAG, more reliable "bundling" has entered the era of mobile interconnection, and the Internet is getting closer and closer to our daily life. while we enjoy the convenience of the network, it also brings the interactive processing of massive data to the network, which puts forward higher requirements for the bandwidth and reliability of the network.

However, the traditional link aggregation technology is limited to one-to-one between two devices, and can not achieve link aggregation between one-to-many devices.

Thus, in order to provide a more reliable network, MC-LAG (Multi-Chassis Link Aggregation Group, Cross-device Link aggregation Group) came into being.

When an access device (which can be a server or a switch) is docked with the upper two network devices An and B, MC-LAG technology can be used to form a cross-device link aggregation group.

The basic idea of MC-LAG is to make two network devices An and B aggregate links with access devices in the same state. In the view of access devices, it is like establishing a link aggregation relationship with the same network device. In this way, the link aggregation technology is extended from one-to-one device docking to the ability to access two opposite devices at the same time, and forms a dual-active system.

Let's take a look at how this dual-active system works.

MC-LAG work process before you understand the MC-LAG work process, learn some basic concepts involved in MC-LAG technology.

DFS Group is the dynamic switching Service Group (Dynamic Fabric Service Group), which is mainly used to pair the two network devices (An and B in the figure) that make up the MC-LAG, and synchronize the interface status, table items and other information of the two devices.

In DFS Group, the roles of devices An and B are distinguished between primary and standby. Under normal circumstances, active and standby devices forward business traffic at the same time.

Peer-link is a directly connected layer 2 link between two MC-LAG devices An and B, which is used to negotiate the interaction of messages and the transmission of part of the traffic.

Keepalive is the heartbeat link between two MC-LAG devices, which carries heartbeat data packets. Its main function is to send dual-master detection messages between the main and standby devices, to carry out dual-master detection, and to prevent equipment An and B from being dual-active.

The MC-LAG member interface is the interface that connects the access devices on two network devices An and B.

After understanding the basic concepts of MC-LAG, we learn more about the process of establishing MC-LAG, which includes the following five steps.

After the configuration of the devices at both ends of the MC-LAG is completed, the devices at both ends will send Hello messages regularly through the Peer-link, and the Hello messages carry their respective DFS Group ID, protocol version number, system MAC and other information.

After receiving the Hello message from the peer, determine whether the DFS Group ID of the peer is the same as yourself, and if so, the pairing is successful.

After a successful match, the primary / standby equipment is elected. The election is carried out according to the MC-LAG priority, and the higher priority is dominant; if the MC-LAG priority is the same, the system MAC,MAC of the two devices is smaller.

Synchronization messages are sent between the main and standby devices for information synchronization.

The master / standby device sends heartbeat detection messages through the Keepalive link, which is mainly used for dual-master detection in the event of Peer-link failure.

After completing the above establishment process, MC-LAG can work properly.

MC-LAG traffic forwarding MC-LAG is mainly used in dual-return access scenarios, that is, access-side device C accesses network-side devices An and B through MC-LAG technology. During normal operation, the uplink traffic and downlink traffic are forwarded through devices An and B by load balancing.

If the above network fails, how does MC-LAG keep the network working properly?

Member interface link failure if the MC-LAG member interface fails, for example, the member interface of device B. Device C on the access side senses the failure of device B member interface and sends all uplink traffic to device A, which is forwarded by device A.

Device B receives traffic from the network side to device C on the access side and forwards the traffic to the working device A to device C on the access side through Peer-link.

MC-LAG device failure if a MC-LAG device fails, for example, device B fails. If device B cannot forward traffic at this time, all traffic is forwarded by device A.

Peer-link failure if Peer-link failure occurs, device An and B cannot forward traffic at the same time, otherwise it will lead to a series of problems such as broadcast storm and MAC drift, so only one device is allowed to forward traffic.

At this point, the standby device of MC-LAG (in this case, device B) will Error-down all its physical interfaces except the Peer-link interface and the management network interface. At this point, all traffic is forwarded only through the MC-LAG master.

Conclusion through the previous introduction, we can see that MC-LAG technology has more advantages over the traditional link aggregation technology, which not only enhances the network reliability, but also simplifies the networking, and realizes the device-level highly available redundancy protection and multi-path forwarding.

In addition, the two MC-LAG network devices operate independently and can be upgraded separately. As long as the upgrade process ensures the normal operation of one device, it will have little impact on the running business.

At present, MC-LAG technology is widely used in the new IP man and cloud data center. While using Spine-Leaf network architecture, MC-LAG can be deployed to ensure the reliability of the network.

In the evolution of 5G, MC-LAG technology will provide a more reliable guarantee for IP bearer network.

This article comes from the official account of Wechat: ZTE document (ID:ztedoc), author: ZTE document

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.