This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>

If you need help, please click here:
MENU

KunLun Charts New Digital Transformation Markets

By Wang Zhen, General Manager for Huawei High-End Servers

The Huawei KunLun server is the world’s first x86-based 32-socket (32S) mission-critical computing platform. It offers the advantages of a groundbreaking CPU interconnect architecture and advanced Reliability, Availability, and Serviceability (RAS) features to meet the ultimate performance and reliability requirements of the most demanding workloads. Since its launch in March 2016, KunLun has been selected and trusted by global customers across industries including finance, government, retail, healthcare, education, and electric power. KunLun has been widely implemented for critical application scenarios such as core databases, mission-critical in-memory computing, and service consolidation.

For these service scenarios, the core databases touch the most important part of the customers’ business systems and therefore are viewed as the most valuable. Many core systems used to be deployed on UNIX servers and, due to the closed UNIX architecture, the entire platform is built on a single vendor’s hardware and software. This traditional deployment leads to poor system compatibility, costly upgrades, and expensive maintenance. The Industrial and Commercial Bank of China (ICBC) has found a paradigm to shift these challenges into opportunities for growth. ICBC has decided to build a new-generation platform for providing its core systems. The defining factors for this new platform are high reliability and performance to be delivered based on an open architecture. Among all of the vast tests and verification performed by ICBC, KunLun servers stand out by meeting all of ICBC’s stringent demands.

For example, on the reliability front, KunLun incorporates proactive fault prediction and analysis mechanisms, which allow the system to generate risk prewarnings and isolate a potentially faulty component before it fails. This feature covers key components such as processors, Dual In-line Memory Modules (DIMMs), and the Node Controller (NC) chip. Leveraging this feature, KunLun delivers up to 99.999 percent reliability and is able to provide 24/7 service assurance for the bank’s core production system. On the performance front, KunLun provides even more advantages with its CPU interconnect architecture. By supporting up to 32 processors, KunLun far outperforms its counterparts with a higher number of transactions per minute. Finally, KunLun wins the trust of the customer by its superior performance, exceptional scalability, and unrivaled reliability, and helps ICBC migrate its mission-critical systems from a closed architecture to an open platform, assuring ICBC stays on track towards digital transformation.

The technical challenges experienced by ICBC is actually a common situation confronting many customers in various industries, including the Postal Savings Bank of China (PSBC), Henan Traffic Police Corps, Heilan Home, and Fujian Medical University Union Hospital. More and more customers are turning to an open architecture, and KunLun emerges as a solution of choice to help customers best resolve their UNIX-to-x86 migration issues. In addition to its superb performance and reliability, more importantly, KunLun provides an open ecosystem. A vast number of Independent Software Vendors (ISVs) from inside and outside China have expanded the ecosystem and developed flexible industry solutions around KunLun. The ISVs include Nantian Electronic Information Co., Ltd., Adtec, Hundsun Technologies Inc., and India’s Infosys. These solutions have been integral to the customers’ success in migrating their critical applications from UNIX servers to the KunLun-based open platform. Customers are now empowered with new levels of agility and rapidity in deploying and upgrading core applications, as well as the ability to lower their Total Cost of Ownership (TCO).

In addition to the core database application scenario, the KunLun-based SAP HANA In-memory Computing Solution has also received much recognition from customers worldwide. A case in point is El Corte Inglés (ECI), a Spain-based mega department store. ECI was planning to deploy a data mart project. As a subset of its data warehouse, data mart will enable ECI to rapidly analyze massive volumes of retail data, keeping ECI better informed of the ever-changing business landscape, meanwhile simplifying O&M for ECI. After methodical observations and evaluation, ECI found that the industry’s popular scale-out HANA solution supports up to only 4 TB of memory, which would fall short on service demands, would necessitate data table division after system expansion, and would further increase O&M workloads. To fulfill ECI’s requirements, the Huawei KunLun SAP HANA solution supports scale-up and, as the world’s first SAP HANA solution supporting 8 TB of Online Analytical Processing (OLAP) capabilities, delivers 30 percent higher performance than the scale-out SAP HANA. KunLun’s scale-up architecture also means that all data is stored on a single node, which eliminates the need for data re-distribution after capacity expansion and therefore drastically simplifies O&M for ECI.

Huawei has worked with global partners to leverage KunLun servers to help customers solve the challenges arising from digital transformation. Looking ahead, Huawei will continue extensive investments in R&D for KunLun. Development of more features for the KunLun server are underway, including logical partitioning, uneven physical partitioning, and I/O expansion enclosures, which are to be launched in 2018. These efforts will augment KunLun to become an even more powerful, reliable x86 server for mission-critical systems that help transition customers smoothly and successfully through their digital transformation journey.

ICT Insights on the go

Scan for mobile reading

KunLun cover en

KunLun Extra

Navigating Critical Enterprise Business Through Digital Transformation

Current catalogue

Comment

Industry Insight

Partner Testimony

Solution

Ecosystem