Cloud Management Deployment: Supporting the Rapid Expansion of Huawei Device Stores

Due to the rapid growth of its consumer business, Huawei has rapidly expanded the construction of its retail stores. Since 2014, Huawei has established more than 300 device stores every year on average. To set up the device store’s network in traditional network deployment mode, the entire set-up process, including planning, deployment, network optimization, and acceptance, takes at least one week. Engineers are also required to perform onsite commissioning multiple times during network deployment. This conventional approach is both inefficient and labor-intensive, failing to support the rapid expansion of Huawei’s device stores.

To address this issue, Huawei uses a cloud management solution to quickly deploy networks for device stores. As shown in Figure 11.4, Huawei’s cloud management solution implements network planning, deployment, optimization, and inspection all on the cloud. By using the cloud management solution, the time required to deploy a device store’s network is reduced from one week to one day. In addition, all devices used in the solution are plug-and-play, requiring no onsite commissioning by engineers. This approach ensures that network services are available on demand during rapid business expansion.

As shown in Figure 11.5, during cloud management deployment, the network administrator can directly scan the barcodes of network devices using a mobile app (CloudCampus APP). This approach greatly facilitates network deployment and effectively supports the rapid expansion of Huawei’s device stores. The cloud management deployment procedure is as follows:

• 1. The tenant administrator imports the ESNs of network devices (APs in this scenario) for stores in batches and then plans offline configurations on the cloud management platform.
• 2. Installation engineers connect and power on APs in device stores. They then log in to the CloudCampus APP and use the barcode scanning function to establish links between APs and the cloud

FIGURE 11.4 Cloud management used by Huawei to quickly deploy networks for device stores.

management platform, and deliver public network configurations to APs through the local management SSID. In this way, APs can be successfully discovered and managed by the cloud management platform.

3. APs periodically report performance data to the cloud management platform. Based on this data, the administrator can perform routine maintenance, periodic inspection, and troubleshooting on devices in device stores using the cloud management platform.

Wi-Fi and IoT Convergence: Helping Huawei Campus Network Go Digital

Huawei is a typical multinational corporation that possesses a large number of high-value assets such as lab instruments, network devices, and servers, as well as common assets, such as office computers, desks, and chairs.

FIGURE 11.5 Deployment by scanning barcodes using a mobile app.

Rough statistics have shown that Huawei possesses more than 400000 assets in total, including over 13000 instruments and meters as well as 120000 electronic devices within the R&D department alone, as shown in Figure 11.6. Effectively managing and utilizing these fixed assets from around the world is a major challenge for Huawei to address.

Since 2014, Huawei has used radio frequency identification (RFID) tags to identify and manage various assets; however, the widespread use of RFID has posed many challenges. For example, it requires the deployment of a large number of RFID tag readers to manage assets. These card readers needed separate power supplies and network cables, leading to a relatively low management efficiency. A management system was also absent for these readers; therefore, it was impossible to detect reader faults quickly. In this case, the relevant administrators did not know whether an RFID reader was faulty or had been relocated until an asset report was

FIGURE 11.6 Examples of Huawei’s typical assets.

generated every month. In general, if an RFID reader was faulty or relocated, its data were not found in the monthly asset report.

Fluawei IT addressed this problem by using its global Wi-Fi network to manage assets. In addition, Huawei IT also developed an asset management solution that converges Wi-Fi and IoT networks, as shown in Figure 11.7.

In this solution, an AP integrates an RFID card to provide asset identification and management functions. It sends asset information back to the asset management platform through the Wi-Fi network. The asset management system then clearly displays the list and status of each person’s key assets in real time, greatly improving asset management efficiency. The biggest advantage of this solution is that it eliminates the need for separate IoT networks; instead, all IoT terminals are managed through the Wi-Fi network, greatly reducing network construction and O&M costs.

The asset management solution proved highly efficient on Huawei’s campus network, being quickly applied to various other scenarios. For example, it was used to manage the IoT sensors embedded in conference rooms, vehicles, campus security devices, meters, and even garbage cans and manhole covers. In this case, the solution connects the IoT sensors to an intelligent operation center through the outdoor Wi-Fi network to achieve ubiquitous connections throughout the campus and ultimately improve the digital level of Huawei’s campus network. Using this solution, Huawei’s campus network becomes a true cornerstone with all-scenario awareness and always-on service continuity.

FIGURE 11.7 Networking architecture of the asset management solution with Wi-Fi and IoT convergence.

Intelligent O&M: Driving Huawei IT to Build a Proactive O&M System

Huawei has designed and developed an intelligent O&M solution that has been used and proven by Huawei IT.

Large-scale Wi-Fi adoption brings unprecedented convenience to daily office work. At the same time, it also leads to more O&M problems such as a large number of faults and difficult fault locating. In the past, Huawei IT received approximately 400 help-seeking calls every month on average, 40% of which were related to wireless networks. To address this, Huawei IT collaborated with the SDN controller product team to find innovative ways to improve Wi-Fi network O&M and fault locating. The two sides optimized the Wi-Fi network O&M algorithm model based on real service data for the field-proven intelligent analysis engine of the SDN controller. Specifically, they summarized four typical Wi-Fi network issue types (connection, air interface performance, roaming, and device issues) from fault data, defined the data criteria for each issue type, and specified how to design intelligent O&M algorithms based on these data criteria.

The comprehensive O&M scenarios on Huawei’s campus network promote improvement on the O&M algorithms of the SDN controller. This, in turn, helps Huawei build a proactive O&M system for campus networks. Specifically, a visualized quality evaluation system is used to constantly drive Huawei IT to proactively improve network quality. For example, the controller dashboard can display severe latency and packet loss rate occurring on an employee’s terminal, and the terminal-associated AP and its wireless signal coverage status can both be viewed based on the AP deployment diagram. If the problem is caused by poor signal coverage, administrators can then make corresponding adjustments in real time. In addition, Huawei’s O&M solution helps achieve automatic fault identification, root cause location, and potential exception prediction, radically changing the passive response of traditional O&M, as shown in Figure 11.8. In the first year after the introduction of the intelligent

FIGURE 11.8 Intelligent O&M scenarios on Huawei’s campus network.

O&M solution, the number of network fault tickets received by Huawei IT decreased by 20%. This figure was then reduced year by year.

BRINGING DIGITAL TO EVERY CAMPUS

As a large multinational enterprise, Huawei faces a complex operations environment typified by a large business volume, multiple customer groups, diversified service scenarios, global resource configurations, and localized operations. Despite this challenge, Huawei has successfully implemented digital transformation in various fields, such as R&D, sales, manufacturing, delivery, logistics, and campus operations. From 2015 to 2019, digital transformation drove Huawei’s rapid growth in business revenue without significantly overwhelming the company’s existing workforce.

Huawei operates a wide range of business around the world and boasts a diverse portfolio of campus use cases and scenarios, including office, R&D, manufacturing, hotel, education, and logistics. Therefore, Huawei’s own campus network is the mold for an intent-driven campus network.

In the past 10 years, Huawei has been using its own campus as a test field and has been building a smart campus featuring “security and controllability, simplified experience, lean cost management, and operational excellence”, as shown in Figure 11.9. This process has led to the development of a series of mature campus network solutions, which have been

FIGURE 11.9 Campus evolution from the digital era to the intelligence era.

rewarding in Huawei’s complex campus service environments. Since their debut, these solutions have been quickly snapped up by customers who want to accelerate their digital transformation and drive business profits.

For the foreseeable future, Huawei will continue to leverage its product portfolio and capitalize on technologies such as big data, IoT, cloud computing, and AI to drive the evolution from digital to intelligent campuses. Throughout this process, Huawei will still adhere to the principle that “we should always test new campus network innovations before releasing them”. It is this basic principle that pushes Huawei to work closely with industry partners to build a win-win campus network ecosystem and provide customers with more mature scenario-specific campus network solutions. As such, Huawei’s ultimate goal is to bring digital to every campus.