Partnership To Co-Innovate Hastens Customer Digital Transformation

0
1032

Since establishing a partnership with SAP in 2012 to fuel joint innovation, Huawei emerged as the first global technology entity to partner with SAP in China.

Fast-forward to 2020 and Huawei and SAP are still going strong and reshaping a wide range of fields, including innovations in intelligent IoT, Big Data, SAP HANA on HUAWEI CLOUD, and SAP HANA appliance, providing a platform for customer digital transformation that spans multiple industries.

SAP HANA is a next-generation in-memory database platform that provides high-performance online transaction processing (OLTP), allowing you to directly query large amounts of real-time service and transaction data.

SAP HANA is so versatile that it plays different roles in the OLAP scenario. For example, it works as an accelerator for other databases, a data source for report analysis, or an analytical data warehouse.

Traditional data processing is a pain point for many, where data is conventionally read from drives for further computing and analytics. As a result, the processing speed is gradually insufficient for the mass service requirements.

The in-memory computing, however, significantly accelerates data processing by allowing the CPU to directly read data from dynamic random access memory (DRAM). When the CPU reads and analyzes data in-memory, the data processing rate is far higher than the conventional data read on drives. The in-memory data reads eliminate I/O bottlenecks for efficient computing and faster data analytics.

SAP HANA greatly improves database performance thanks to the in-memory read/writes, which are orders of magnitude faster than read/writes on drives. Since the data is stored in memory, the latency for data queries with SAP HANA is slashed to just nanoseconds. Likewise, for data persistence, it ensures that all transactions are committed to the log volume of the drive-in real-time, preventing data loss even in the event of power failure.

With the growing demand for faster, better data storage, the SAP HANA in-memory database is facing the following challenges:

  • Limited capacity: Due to its conventional hardware design, the largest memory offered by the DRAM fails to keep in pace with the ever-growing capacity of the SAP HANA in-memory database.
  • High costs: Because of its price, DRAM drives the total cost of ownership (TCO) for hardware.
  • Prolonged restart: Data loading may last for hours, because data can be accessed only after it is loaded from the persistence layer (drive storage) to the memory.

These problems are addressed by Intel® Optane™ persistent memory (PMem), a next-generation non-volatile dual in-line memory module (DIMM) running on the Intel Optane™ Technology. This technology boosts the storage media to 1,000 times that of the NAND Flash, and a storage density 10 times higher than DRAM, helping turbocharge the storage capacity of the 2nd-gen Intel® Xeon® Scalable processors. Equipped with Intel PMem, the SAP HANA solution provides large storage capacity, non-volatile memory, and high running performance at lower costs.

Huawei was one of the first to support this cutting-edge tech, and subsequently launched the next-generation FusionServer Pro intelligent servers running on the 2nd-gen Intel® Xeon® Scalable processors and PMem. In this product lineup, the FusionServer Pro 2288H V5 (2-socket), FusionServer Pro 2488H V5 (4-socket), and KunLun 9008 V5 (8-socket) are incorporated into the SAP HANA tailored data center integration (TDI) solution. With memory capacity up to 24 TB, this solution builds an end-to-end SAP HANA infrastructure platform for real-time services.

The Huawei SAP HANA equipped with the latest PMem supports three memory capacity specifications: 128 GB, 256 GB, and 512 GB per DIMM. Each processor (configured with 256 GB DRAM and 512 GB PMem) can provide memory capacity up to 4.5 TB, which is three times that of a processor without PMem.

Thanks to the App Direct (AD), a mode unique to the PMem, the data stored in the PMem is persistent. The SAP HANA persistence layer comprises the log volume and data volume. All SAP HANA transactions in the memory are saved to the log volume of the drive-in real-time.

Memory data subjects are periodically and automatically stored in the data volume of the drive. After the SAP HANA system is restarted, its master data stored in the non-volatile memory can still be accessed in the memory without being reloaded from the drive. On top of the data persistence, the restart of the persistent memory is 12.5 times faster than that of traditional memory methods.

The traditional method for expanding service capabilities is scale-out, that is, to deploy more servers for larger memory capacity. In contrast, scale-up is a better option because it reduces the number of servers required and the costs of key components, and streamlines server cluster management. The latest version of PMem offers lower capital expenses (CAPEX) and operating expenses (OPEX) to reduce TCO. In addition, it supports persistent data storage and data recovery within seconds, providing more reliable and secure services.

Over the past 18 years, Huawei has continuously invested in server technologies for technology and product innovation. Huawei is a technological enterprise that digs into customer and industry problems to provide products and solutions tailored to specific scenarios. According to the ITIC 2018 Global Server Hardware, Server OS Reliability Report, the Huawei FusionServer Pro intelligent servers meets enterprise requirements by delivering 99.99% uptime and 99.999% availability.

Huawei’s next-generation servers are designed to meet diversified enterprise requirements in an energy-efficient and cost-effective manner. Huawei servers also provide intelligent management and high energy efficiency to reduce O&M costs.

Learn more about FusionServer Pro.

Do you have a story that you think would interest our readers?
Write to us editorial@cio.co.ke

LEAVE A REPLY

Please enter your comment!
Please enter your name here
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.