Performance Of Solid State Drives Research Articles

Improving flash write performance by using update frequency

Solid-state drives (SSDs) are quickly becoming the default storage medium as the cost of NAND flash memory continues to drop. However, flash memory introduces new challenges, as data cannot be eciently updated in-place. To overcome the technology's limitations, SSDs incorporate a software Flash Translation Layer (FTL) that implements out-of-place updates, typically by storing data in a log-structured fashion. Despite a large number of existing FTL algorithms, SSD performance, predictability, and lifetime remain an issue, especially for the write-intensive workloads specific to database applications. In this paper, we show how to design FTLs that are more efficient by using the I/O write skew to guide data placement on flash memory. We model the relationship between data placement and write performance for basic I/O write patterns and detail the most important concepts of writing to flash memory: i) the trade-o between the extra capacity available and write overhead, ii) the benefit of adapting data placement to write skew, iii) the impact of the cleaning policy, and iv) how to estimate the best achievable write performance for a given I/O workload. Based on the findings of the theoretical model, we propose a new principled data placement algorithm that can be incorporated into existing FTLs. We show the benefits of our data placement algorithm when running micro-benchmarks and real database I/O traces: our data placement algorithm reduces write overhead by 20% - 75% when compared to state-of-art techniques.

Design and Implementation of Remote Destruction SSD

An implementation of a remote destruction solid state drive (SSD) was introduced in the design. The advanced technology of rapid data destruction, the technology of breakdown the flash with high-voltage pulse and the technology of GSM is combined together. As a result, user can operate through a simple command (text message or a call) to destroy all data in SSD, as long as the user confirms SSD is out of control. The method of erasing data and destroying the flash I/O are both executed, so the data can’t be recovered after destruction. At last, some tests are displayed to verify the performance of the remote destruction SSD.

An endurance solution for solid state drives with cache

NAND flash memory has become one of the most popular storage media for portable devices, such as MP3 players, MMC cards and solid state drives (SSDs). Due to erase-before-write characteristics of NAND flash memory, wear-leveling strategy is very important in determining the performance and lifetime of NAND flash memory in solid state drives. In this paper, to prolong the lifetime and improve the performance of SSDs with cache, we propose an effective wear-leveling algorithm based on a novel “triple-pool” design. Comparing with previous wear-leveling algorithms, experimental results show that our algorithm lengthens the lifetime and reduces the write amplification.

Ferroelectric (Fe)-NAND Flash Memory With Batch Write Algorithm and Smart Data Store to the Nonvolatile Page Buffer for Data Center Application High-Speed and Highly Reliable Enterprise Solid-State Drives

A ferroelectric (Fe)-NAND flash memory with a batch write algorithm and a smart data store to the nonvolatile page buffer is proposed. An enterprise solid-state drive (SSD) for a data center is a future promising market of NAND flash memories. The critical problem for such an enterprise SSD is a slow random write. The write unit in a NAND flash memory is a page, 4-8 KBytes. Because the minimum write unit of the operating system is a sector, 512 Bytes, a random write to write a smaller data than a page size frequently happens, which creates a garbage. As a garbage accumulates, a garbage collection is performed to increase a workable memory capacity. The garbage collection takes as much as 100 ms, which is 100 times longer than a page program time, 800 μs, and thus causes a serious performance degradation. In the proposed Fe-NAND flash memory, the data fragmentation in a random write is removed by introducing a batch write algorithm where a page buffer in the Fe-NAND flash memory temporarily stores a program data. The memory cell program starts after the program data as much as the page size accumulates in page buffers. As the data fragmentation is eliminated, the SSD performance can double. In addition, the nonvolatile page buffer realizes a power-outage-immune highly reliable operation. With a low program/erase voltage of 6 V and a high endurance of 100 million cycles, the proposed Fe-NAND flash memory is most suitable for a highly reliable highspeed low-power data-center-application enterprise SSD.