The past few years have seen the rise of solid-state drives (SSDs) in both the consumer and enterprise segments. SSDs are more robust against shock, consume less power, use capacity more efficiently, and offer longer life, faster access times and better read and write performance.
However, like any Flash-based storage device, SSDs can face data and file system corruption in the event of a sudden power loss. Kingston® puts its drives through a very strenuous power cycling test where we intentionally cut power to the drive to analyse what happens to user data and the file system. Suffice it to say that client-level users should be fine in the event of an unexpected power loss to their systems. However, please do not take this as a licence to conduct your own power cycling testing. As with a traditional hard disk drive, data loss may still result depending on what processes the drive was doing the moment it lost power.
When Kingston engineering set out to design its first server-class SSD, we asked our enterprise customers what features they wanted to be implemented and a power failure mechanism was high on the list. An effective power-failure data protection mechanism needs to function before and after a disruptive power failure in order to provide comprehensive data protection.
During a clean shutdown, most host systems initiate a command (the STANDBY IMMEDIATE command) to an SSD to give it time to prepare for the shutdown. This gives the SSD enough time to save data currently in transition (in temporary buffers), LBA mapping information and/or metadata to the non-volatile NAND media.
In the event of an unsafe power failure, the SSD abruptly loses power before the host system can initiate the STANDBY IMMEDIATE command, which can result in the loss of data, mapping information and page corruption.
A well-designed enterprise SSD contains both hardware and firmware safeguards to protect and ensure the integrity of the data and file system. To ensure that the drive has enough power to complete the necessary tasks during an improper shutdown, SSD manufacturers use either aluminium electrolytic supercapacitors or multiple tantalum capacitors. Although both can be very effective, tantalum capacitors offer a superior solution for enterprise SSD usage.
The effectiveness of a supercapacitor wears out over time. The loss rate is also accelerated by higher operating voltages and higher temperatures. Supercaps are said to be “wet” and will dry out over time, especially when exposed to higher than ambient temperatures. With the increased endurance of enterprise-grade NAND used in SSDs, the operating life of the drive can easily exceed that of the supercap’s effective capacitance.
The advantages of discrete tantalum capacitors over supercapacitors are:
Tantalum capacitors are solid/dry (no liquid electrolyte). There is no known wear on our mechanism, meaning that the capacitors can maintain their designed capacitance for decades when used within design limits.
Wide operating temperature range: Tantalum capacitors are able to operate over a very wide temperature range. They are often specified for operating over the range -55C to +125C. This makes them an ideal choice for use in equipment for use in harsh environmental conditions.
Tantalum capacitors have a higher volumetric efficiency (CV/cc) when compared to other types of capacitors. For instance, a 10-microfarad tantalum capacitor can replace a 100-microfarad aluminium capacitor.
In addition to the use of tantalum capacitors on the E100 SSD, a power-failure detection circuit (pFAIL) is employed to monitor the voltage supply to the SSD. If voltage drops below the threshold for the drive, a signal is sent to the SSD controller to begin the process of hardening any data in transit to NAND and safely shut down the drive.