Online Transaction Processing Applications Big Data Software Defined Storage Virtualization Edge Computing
Online Transaction Processing Applications
For flash sales and red envelope rains during 618 and Double 11 Shopping Festivals, the data center has to deal with tens of millions of user requests in a very short time. Its access data volume and access requests are concentrated in a short period of time, which requires the underlying storage system to have stable high performance, high reliability, and low latency. The impact of storage system performance jitter on business can be amplified step by step and even ultimately cause IO timeout. Zhenyue 510 has the built-in efficient QoS management algorithm to greatly reduce inter-IO interference. The accurate media voltage-drifting prediction algorithm and the powerful LDPC algorithm achieve “Error Correction Once Read”, thus greatly reduce performance fluctuation, improve the long-tail latency, and provide the highly consistent performance for the mission-critical applications.
Big Data
Big data analysis is to find critical values in massive data. One of the main characteristics of big data applications is real-time or near real-time, which requires infrastructures to have high performance, high throughput, and large capacity. Zhenyue 510 provides high bandwidth of 14GB/s, large storage capacity of tens of Terabytes, and supports cheap storage media such as QLC, which can support efficient and low-cost big data analysis.
Software Defined Storage
Software defined storage separates the control plane from the data plane of a storage system, enabling the performance to increase linearly with the growing number of nodes, while significantly reducing overall system costs through the use of general-purpose servers. With the rapid evolving software defined storage technology, the number of applications hosted on it continues to increase, while the requirements for IO performance of each application also continue to increase, which all put forward a high demand for the performance and response latency of the underlying SSD. Zhenyue 510 provides a PCIe Gen5 interface with IO processing capability of 3400 Kilo IOs per second, which can fully release the performance of the underlying flash media and improve the performance of the entire system.
Cloud computing is another innovation in the information age after the Internet and computers. In a cloud environment, computing, network, and storage exist in the form of pooled resources through virtualization technology, users apply and pay as needed, so as to achieve high flexibility, scalability, and cost-effectiveness. Zhenyue 510 provides up to 32 pure hardware implemented Virtual Functions. A physical device can be virtualized into 32 devices that can be directly used by virtual machines, greatly reducing the performance overhead of virtualization. Zhenyue 510 can effectively reduce performance interference among various tenants through the self-developed virtualization performance management mechanism.
Edge Computing
The Internet of Things (IoT) has given rise to edge computing. Data generated by massive terminal device will be collected and processed in the edge data center, thereby reducing the storage, bandwidth, and computing pressure of the cloud data center. The computing latency can be reduced as the edge computing node is closer to users, bringing better business experience. Edge data centers need to build high throughput and reliable computing systems, placing high demands on bandwidth and reliability of SSDs. Zhenyue 510 provides high bandwidth of 14GB/s and error correction capability of 10^-18, giving the edge data center better system performance and higher data reliability.