Flash storage/NVMe (Non-Volatile Memory Express) is a scalable, high performance CPU PCI-E Gen3 direct connect to NVMe devices; designed for Client and Enterprise server systems using Solid State Drive (SSD) technology that was developed to reduce latency and provide faster CPU to data storage device performance.
By deploying NVMe, our customers can benefit from reduced latency, increased Input/Output Operations Per Second (IOPS), and lower power consumption. This means that customer solutions can perform more work in less time, translating into lower costs and increased revenues. Thus the ROI benefits for customers from NVMe enabled servers is immediate and substantial. Supports standard NVMe drivers.
The capacity of flash storage systems will continue to rise: the first flash modules of 4 terabytes (TB) in only 2.5 inches are appearing. Which is six times better than the most recent record. And the price / capacity ratio will continue to improve. Prices will fall further: in 2018, the price of one terabyte of flash memory will be the same than conventional hard disks. The “100% Flash” approach is already proving more economical. Studies show that, as of now, the overall cost of SSDs, including operation and maintenance, is less than that of conventional discs.
|Measurement||NVME SSD||SATA3/SAS3 SSD||NVMe Improvement|
|Bandwidth vs. SATA3||128K Sequential Read||2788.77 MB/s||470.60 MB/s||5.93x|
|128K Sequential Write||1838.98 MB/s||453.33 MB/s||4.06x|
|Bandwidth vs. SAS3||128K Sequential Read||2788.77 MB/s||980.34 MB/s||2.84x|
|128K Sequential Write||1838.98 MB/s||421.25 MB/s||4.37x|
|Latency vs. SATA3||4K Sequential Read||15.9 µs||47.3 µs||2.97x|
|4K Sequential Write||17.6 µs||51.4 µs||2.92x|
|Latency vs. SAS3||4K Sequential Read||15.9 µs||114.1 µs||7.18x|
|4K Sequential Write||17.6 µs||109.853 µs||6.24x|
For the bandwidth test, Supermicro ran the IOMeter benchmarking tool on two SYS-1028U-TNR4T+ Ultra 1U servers, the first with standard SAS3 and SATA3 SSDs, and the second with NVMe SSDs. The upgraded configuration with NVMe delivered a maximum of 5.9x the performance of the server with standard SSDs.
For latency, the table shows the upgraded configuration with NVMe performed with up to 7.2x improved latency compared to the servers with standard SSDs.
While the base configuration delivered excellent performance, the dramatic performance improvement offered with the NVMe SSDs makes this upgrade an excellent investment for any business looking to improve the performance of its IT.
Flash Storage/NVMe leverages the experience gained with known block storage access and unlocks the performance capabilities of newer and faster devices. It can be implemented in a variety of topologies and configurations, from dedicated direct-attached storage (DAS) on a server or storage system (e.g., back end) or as a shared front-end alternative to block storage protocols.
While initially being deployed as DAS back-end storage in servers or storage systems accessing fast flash storage, NVMe will eventually be used in other ways. The SCSI command set, for example, gradually found its way from SAS to iSCSI, Fibre Channel Protocol (FCP), and others, both on the back end to access hard disk drives (HDDs) or SSDs, as well as on the front end for servers to access storage systems.
While SATA allowed for one command queue capable of holding 32 commands, NVMe enables 65,536 (64K) queues with 64K commands per queue. As a result, the storage I/O capabilities of flash can now be fed across PCIe much faster, enabling modern multi-core processors to complete more useful work in less time.
Like the robust PCIe physical server storage I/O interface it leverages (e.g., the data highway), NVMe provides both flexibility and compatibility, while removing complexity, overhead, and latency. Some environments, however, might phase in emerging external NVMe-based shared storage systems. Over time we can also expect to see NVMe being deployed across more servers and inside storage systems (or appliances) to access fast NVM flash-based storage.