Industrial solid state drives (SSDs) based on NAND flash memory have the clear advantage for avoiding expensive downtime when built in memory fails: They have no moving parts, making them more resistant to shock and vibration; they can withstand harsh thermal, mechanical, and electrical conditions; they’re faster, smaller, and can be continually written with data; and many offer detection mechanisms like error correction code (ECC) and robust flash management techniques such as wear-leveling, bad block remapping, and write abort handling.

But choosing the right device gets a little more complex depending on specific budget and application requirements. A recent white paper from Cactus Technologies, “Rugged flash memory: Anatomy of an industrial storage device,” discusses the benefits and drawbacks of a number of different features.

For example, finer node geometries might offer more capacity at a lower price, but are more prone to errors and power leakage. The number of bits stored per cell will also impact the longevity of a device; SLC chips offer better endurance compared to cheaper MLC and TLC chips, which are better suited to less write-intensive applications. Meanwhile, ECC hardware is only as good as the accompanying controller architecture and firmware algorithm. And finally, performance means nothing if the device can’t stand up to the environmental demands of the application. Does it need a die-cast metal chassis? Does it need to operate in extreme temperatures? There’s a lot to mull over.

If you’re a little confused about what to look for when it comes to flash memory cells, flash controller architecture, and rugged flash construction to achieve the right performance, endurance, and price point, check out the white paper at http://embedded-computing.com/white-papers/white-industrial-storage-device/.