Yunze Nand Storage Solutions for Data Center Applications

A staggering amount of cloud applications, each with an ever-increasing individual complexity is covered by one-size-fits-all SSD solutions provided by the mainstream market. Under real world workloads, a standard Data Center SSD degrades by 50-75% over a period of 4-10 months. Cloud service providers are confronted with significant reduction of performance and increased latency that do not meet the service levels required by end customers. Additionally increased cost for early replacement of the storage SSD and additional service costs reduces the profitability of datacenters. Hence, there is significant potential to improve efficiency.

	Mainstream market SSD	Yunze Data Center SSD
Performance	Optimized for FoB peak performance Inconsistent performance over the lifetime of the SSD	2x - 5x times better consistent steady state performance over the entire lifetime of the SSD
Endurance	Datasheet endurance not achievable in real life conditions	2x greater SSD endurance compared to mainstream-market solutions
Latency	Unpredictable latency behavior with significant spikes impacting application responsiveness	Consistent low latency resulting in reliable and predictable service levels
Use case	One-size-fits-all, emulation of HDD benchmark tests	Optimized for real-world workloads in cloud applications

Big Data

Deep analysis of captured and stored data requires high amount of fast storage with parallel operation of writes and reads. The data size depends highly of the data source.

Yunze-Video Streaming

Video streaming is a steadily growing business replacing traditional TV. Massive amounts of data is written once and read with multiple streams in parallel.

Machine Learning

Machine Learning takes a multitude of training data samples to build an optimized model that can be used to identify conditions and decide on actions.

Artificial Intelligence

AI or artificial intelligence is closely connected to machine learning and covers the analysis of data, the evaluation of scenarios and the resulting reactions.

Internet of Things

The Internet of Things stands for a myriad of different applications. Data Centers are used at the edge where preprocessed data needs to be stored for further analysis or documentation.

Computational Storage

Computational storage shall offload processing tasks from the CPU & perform parallel operations within the storage. This data processing requires dedicated computing elements within the storage device.