How to Design a Reliable Product Using NAND Flash Memory

Users of managed flash devices¹ asked our applications team to recommend a process for conducting an accurate NAND flash memory lifetime reliability analysis.  The results from this analysis, when properly conducted, can provide confidence to users that their operating systems are not corrupted and data stored in the NAND flash memory will maintain its integrity over a long lifetime.  This means that their photos, videos, music, contacts, social media activity, confidential information, etc., will be available and accessible whenever they need it, even if the data was captured a while back.

Our team developed a process to properly perform such an analysis that we recently published, and titled, “Analyzing Managed Flash Device Lifetime Reliability.” It’s a four-part series of short tech briefs that present the steps and calculations required to perform an accurate lifetime reliability analysis, as follows:

• Check and calculate the total lifetime P/E cycles                       Part 1 available here
• Calculate data retention for a worst case scenario                   Part 2 available here
• Improve data retention using refresh functionality                 Part 3 available here
• Conduct the lifetime reliability analysis                                      Part 4 available here

Here’s a quick overview:

Check and calculate the total lifetime Program/Erase cycles

The lifetime reliability analysis requires a total Program/Erase (P/E) cycle count.  It’s determined by calculating the total amount of the data programmed (written) into the NAND flash memory.  This is a very important metric as NAND flash memory lifetime is defined by a finite P/E count, so knowing an application’s P/E count helps to predict when the flash memory will wear out.  For our sample analysis, we used a surveillance camera as the device, and estimated workload conditions in an early development phase of the managed flash storage design.  The total P/E cycle count was calculated as follows:

Though these results are rough and estimated in an early development phase, they can be used as a foundational comparison throughout the lifetime reliability analysis process. 

Calculate data retention for a worst case scenario

The lifetime reliability analysis also requires a calculation of the data retention time.  In order to do this, the thermal profile of the device needs to be obtained.  The thermal profile is typically selectable or customizable by users so they can request data for various temperatures.  Since higher temperatures cause data loss, it is important to obtain the NAND flash thermal profile, which is typically available from the NAND flash memory supplier.

Once the thermal profile is determined, the worst case data retention can be calculated.  This is where the total P/E cycle count is used, as well as the ambient temperature of the surveillance camera.  We calculated the worst case data retention as follows:

The worst case data retention has a total P/E cycle count of 1,460 after 10 years in conjunction with the custom system thermal profiles represented.

Improve data retention using refresh functionality

Now that the total P/E cycle count and worst case data retention have been calculated, the two main NAND flash memory reliability parameters are available to complete the lifetime reliability analysis.  These values can be compared to the managed flash device’s specifications as a means to initially foresee potential issues.  If the calculated data retention does not meet the requirements of the managed flash device design, refresh functionality can be used to improve data retention (see link).

Final thoughts

The “Analyzing Managed Flash Device Lifetime Reliability” series that we published helps managed flash device users gain a deeper understanding of the reasoning used to develop a robust lifetime reliability analysis, and the steps required to conduct that study.;

General information for KIOXIA memory products is available here.

NOTES:

¹ A managed flash device combines raw NAND flash memory and an intelligent controller in one integrated package, enabling internal memory management. 

²Definition of capacity - KIOXIA Corporation defines a kilobyte (KB) as 1,024 bytes, a megabyte (MB) as 1,000,000 bytes, a gigabyte (GB) as 1,000,000,000 bytes, a terabyte (TB) as 1,000,000,000,000 bytes and a petabyte (PB) as 1,000,000,000,000,000 bytes. A computer operating system, however, reports storage capacity using powers of 2 for the definition of 1Gbit = 2³⁰ bits = 1,073,741,824 bits, 1GB = 2³⁰ bytes = 1,073,741,824 bytes, 1TB = 2⁴⁰ bytes = 1,099,511,627,776 bytes and 1PB = 2⁴⁰ bytes = 1,125,899,906,842,624 bytes and therefore shows less storage capacity. Available storage capacity (including examples of various media files) will vary based on file size, formatting, settings, software and operating system, and/or pre-installed software applications, or media content. Actual formatted capacity may vary.

TRADEMARKS:

All other company names, product names and service names may be trademarks of third-party companies.

DISCLAIMERS:

KIOXIA America, Inc. may make changes to specifications and product descriptions at any time.  The information presented in this blog is for informational purposes only and may contain technical inaccuracies, omissions and typographical errors.  Any performance tests and ratings are measured using systems that reflect the approximate performance of KIOXIA America, Inc. products as measured by those tests. In no event will KIOXIA America, Inc. be liable to any person for any direct, indirect, special or other consequential damages arising from the use of any information contained herein, even if KIOXIA America, Inc. are advised of the possibility of such damages.