Figure 6 - uploaded by Bo Chen
Content may be subject to copyright.
Comparison of I/O throughput between OpenNFM and RedFlash, obtained from fio benchmark under different read/write patterns. SR: sequential read; RR: random read; SW: sequential write; RW: random write.
Source publication
Flash memory has been used extensively as external storage of smartphones, tablets, IoT devices, laptops, etc. Therefore, more and more sensitive or even mission critical data are stored in flash and, once the data turn obsolete, securely deleting them is necessary for both regulation compliance and privacy protection. Traditional secure deletion o...
Context in source publication
Context 1
... threshold for wear leveling was set as 10 (i.e., when the difference of P/E cycles of two blocks exceeds 10, wear leveling is triggered) which can ensure good wear leveling effectiveness. The results are shown in Figure 6. The benchmark evaluated different I/O patterns including sequential read/write and random read/write. ...
Citations
... These can be, e.g., solutions that determine data quality, or similarities (Long et al., 2020) to data already marked for deletion. In particular, duplicates (Chen and Chen, 2022;Rashid et al., 2012;Pachpor and Prasad, 2018) and sensitive data (Pecherle et al., 2011) are often mentioned in our corpus for deletion. Languages include anything that allows data to be identified by giving a set of instructions according to a given grammar. ...
Data is an important asset and managing it effectively and appropriately can give companies a competitive advantage. Therefore, it should be assumed that data engineering considers and improves all phases of the data life cycle. However, data deletion does not seem to be prominent in theory and practice. We believe this is for two reasons. First, the added value in deleting data is not always immediately apparent or has a noticeable effect. Second, to the best of our knowledge, there is a lack of structured elaboration on the topic of data deletion that provides a more holistic perspective on the issue and makes the topic approachable to a greater audience. In this paper, an extensive systematic literature review is conducted to explore the topic of data deletion. Based on this, we present a data deletion taxonomy to organize the subject area and to further professionalize data deletion as part of data engineering. The results are expected to help both researchers and practitioners to address the end of the data life cycle in a more structured way.
Sanitization is an effective approach for ensuring data security through scrubbing invalid but sensitive data pages, with the cost of impacts on storage performance due to moving out valid pages from the sanitization-required wordline, which is a logical read/write unit and consists of multiple pages in high-density SSDs. To minimize the impacts on I/O latency and data security, this paper proposes a polling-based scheduling approach for data sanitization in high-density SSDs. Our method polls a specific SSD channel for completing data sanitization at the block granularity, meanwhile other channels can still service I/O requests. Furthermore, our method assigns a low priority to the blocks that are more likely to have future adjacent page invalidations inside sanitization-required wordlines, while selecting the sanitization block, to minimize the negative impacts of moving valid pages. Through a series of emulation experiments on several disk traces of real-world applications, we show that our proposal can decrease the negative effects of data sanitization in terms of the risk-performance index, which is a united time metric of I/O responsiveness and the unsafe time interval, by 16.34% on average, compared to related sanitization methods.
