Reliability Specifications

Uncorrectable Bit Error Rate (UBER)

For the enterprise application, JEDEC recommends that UBER shall be below 10-16.

< 1 sector per 10¹⁷ bits read

Rate of occurrence of data errors, equal to the number of data errors per bits read as specified in the JESD218 document of JEDEC standard.

Mean Time Between Failures (MTBF)

By definition, Mean Time between Failures (MTBF) is the estimated time between failures occurring during SSD operation.

2 million hours

Data Retention

Data retention was measured by assuming that SSD reaches the maximum rated endurance at 40C in power-off state.

3 months

Drive Write Per Day (DWPD)

The endurance of SSD in enterprise application is defined as the maximum number of drive writes per day that can meet the requirements specified in the JESD218 document of JEDEC standard.

5 drive writes per day over 5 years

Petabyte Written (PBW)

Relational formula between DWPD and PBW is below: PBW = DWPD x 365 x 5 x User capacity

29.2 PB

Sustained Random Read/Write Performance (IOPS) - Maximum

Random performance was measured by using FIO in Ubuntu with queue depth 32 by 16 workers. Measurements were performed on a full Logical Block Address (LBA) span of the drive in sustained state. Actual performance may vary depending on use conditions and environment.

Random 4KB Read: 750K
Random 4KB Write: 120K
Random 8KB Read: 390K
Random 8KB Write: 60K

Sequential Read/Write Performance -Maximum

Sequential performance was measured by using FIO in Ubuntu with queue depth 32 by 16 worker. Actual performance may vary depending on use conditions and environment.

Sequential 128KB Read MB/s 5,500
Sequential 128KB Write MB/s 1,800

Latency

Typical values.

(sustained state)

Random Read/Write: 90 / 20 μsec
The random read/write latency is measured by using FIO in Ubuntu and 4KB transfer size with queue depth 1 on a random workload of sustained state.
Sequential Read/Write: 90 / 20 μsec
The sequential read/write latency is measured by using FIO in Ubuntu and 4KB transfer size with queue depth 1 on a sequential workload of sustained state.
Drive Ready Time: 2 seconds
The maximum taking time to be ready for receiving commands after power-up (CSTS.Ready=1). It is expected that I/O commands may not be completed at this point.

Quality of Service (QoS)

QoS is measured using FIO (99/99.99 %) with queue depth 1 and 128 on 4 KB random and write.

Quality of Service (99%)
- Read (4KB) μsec
  
  QD=1 95, QD=128435
- Write (4KB) μsec
  
  QD=1 60, QD=128 5795
- QoS is measured as the maximum round-trip time taken for 99 % of commands to host.
Quality of Service (99.99%)
- Read (4KB) μsec
  
  QD=1 140, QD=128 565
- Write (4KB) μsec
  
  QD=1 200, QD=128 9540
  QoS is measured as the maximum round-trip time taken for 99.99 % of commands to host.

IOPS Consistency

IOPS consistency measured using FIO with queue depth 128.

IOPS Consistency (%) = (IOPS in the 99.9% slowest 1-second interval)/(average IOPS during the test).

Random Read (4 KB) % 97
Random Write (4 KB) % 94
Random Read (8 KB) % 94
Random Write (8 KB) % 92

Out of Band Management (SMBUS)

Provides out-of-band management by means of SMBUS interface. This requires 3.3V auxiliary voltage.

SMBUS access includes the VPD page and temperature sensor.

Hot Plug Support

Supports PCIe presence detect and link-up detect

Related Information