A Hotness Filter of Files for Reliable Non-Volatile Memory Systems

Wu, Chin-Hsien; Wu, Po-Han; Chen, Kuo-Long; Chang, Wen-Yen; Lai, Kun-Cheng

doi:10.1109/tdsc.2015.2399313

Cited by 5 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As mentioned before, the limited endurance of NVM devices may cause some cells or blocks/pages to wear out faster than others, since some data may be updated much more frequently than others [61], [118], [135]. This results in permanent loss of these blocks, reduced storage capacity and possibly in file corruption.…”

Section: ) Endurancementioning

confidence: 99%

“…Wu et al [134], [135] use two different heuristics to identify if frequently updated (hot) files are present. For instance, files that are marked as read-only, files belonging to specific users that rarely log into the system, read-only operating system files and files that are rarely accessed or are related to processes that rarely run, are good candidates for cold or frequently read files.…”

Section: ) Endurancementioning

confidence: 99%

See 1 more Smart Citation

Non-Volatile Memory File Systems: A Survey

et al. 2019

View full text Add to dashboard Cite

For decades, computer architectures have treated memory and storage as separate entities. Nowadays, we watch the emergence of new memory technologies that promise to significantly change the landscape of memory systems. Byte-addressable non-volatile memory (NVM) technologies are expected to offer access latency close to that of dynamic random access memory and capacity suited for storage, resulting in storage-class memory. However, they also present some limitations, such as limited endurance and asymmetric read and write latency. Furthermore, adjusting the current hardware and software architectures to embrace these new memories in all their potential is proving to be a challenge in itself. In this paper, recent studies are analyzed to map the state-of-the-art of NVM file systems research. To achieve this goal, over 100 studies related to NVM systems were selected, analyzed, and categorized according to their topics and contributions. From the information extracted from these papers, we derive the main concerns and challenges currently being studied and discussed in the academia and industry, as well as the trends and solutions being proposed to address them.

show abstract

Section: ) Endurancementioning

confidence: 99%

Section: ) Endurancementioning

confidence: 99%

Non-Volatile Memory File Systems: A Survey

et al. 2019

View full text Add to dashboard Cite

show abstract

“…The primary goal of these works is to increase I/O bandwidth with multi channel architectures, enabling the interleaving of data over multiple flash chips. Reference [16] augments this with a hot-cold data identification technique that improves endurance of the system. However, these techniques do not provide mechanisms to protect data in case of failure, or guarantee the upper bound of I/O access latency.…”

Section: B Related Workmentioning

confidence: 99%

A Real-Time Dependable Flash Storage System

McEwan

Komsul

2019

IEEE Access

View full text Add to dashboard Cite

One of the limitations of flash memory in real-time and high dependability systems is its need for garbage collection, resulting in performance degradation due to non-deterministic response times. Recent work has presented RAID architectures for solid state storage systems. These RAID architectures increase the dependability from a data storage perspective but they do not provide application level dependability when real-time response times are required. In this study we present a garbage collection aware Flash Translation Layer that offers guaranteed access time to a solid state RAID array by managing incoming requests and preventing them from being blocked by ongoing garbage collection. We present a novel serial technique with a dynamic page allocation mechanism that eliminates non-deterministic behaviours of the garbage collectors in the array. The result is real-time access guarantees that maintain the data dependability enhancements using a run time parity migration technique. The mechanisms are evaluated using a trace driven simulator and a number of synthetic and realistic traces. Simulation results indicate that the garbage collection aware techniques offer improved upper bound response times for I/O requests of up to 73% compared to an existing mechanism, without disturbing the data dependability at the storage level. Traces dominated by random writes exhibit similarly significant enhancements.INDEX TERMS Data storage systems, flash memories, fault tolerance, flash translation layer, garbage collection, real-time systems, SSD RAID.

show abstract

“…The past two decades witness considerable efforts on studies of HDI from direct scheme [6], [7] to indirect scheme [5], [8]- [12]. The HDC [6] and two-level LRU [7] analyze the access behaviors by directly recording the statistical rule of logical block addresses (LBAs).…”

mentioning

confidence: 99%

“…In summary, all these HDI schemes focus on how to effectively capture the frequency and recency information of write accesses with less runtime consumptions and memory overheads. However, LBA accesses tend to be stochastic and time-variant [5], [10]- [12]. It is challenging to precisely identify the hot ratio of all kinds of workloads even for the recent state-of-the-art HDI schemes.…”

mentioning

confidence: 99%

Self-Learning Hot Data Prediction: Where Echo State Network Meets NAND Flash Memories

Luo

Fang

Sun

et al. 2020

IEEE Trans. Circuits Syst. I

View full text Add to dashboard Cite

Well understanding the access behavior of hot data is significant for NAND flash memory due to its crucial impact on the efficiency of garbage collection (GC) and wear leveling (WL), which respectively dominate the performance and life span of SSD. Generally, both GC and WL rely greatly on the recognition accuracy of hot data identification (HDI). However, in this paper, the first time we propose a novel concept of hot data prediction (HDP), where the conventional HDI becomes unnecessary. First, we develop a hybrid optimized echo state network (HOESN), where sufficiently unbiased and continuously shrunk output weights are learnt by a sparse regression based on L 2 and L 1/2 regularization. Second, quantum-behaved particle swarm optimization (QPSO) is employed to compute reservoir parameters (i.e., global scaling factor, reservoir size, scaling coefficient and sparsity degree) for further improving prediction accuracy and reliability. Third, in the test on a chaotic benchmark (Rossler), the HOESN performs better than those of six recent state-of-the-art methods. Finally, simulation results about six typical metrics tested on five real disk workloads and on-chip experiment outcomes verified from an actual SSD prototype indicate that our HOESN-based HDP can reliably promote the access performance and endurance of NAND flash memories.Index Terms-NAND flash memory, solid state disk (SSD), echo state network (ESN), hot data prediction, regularization.

show abstract

A Hotness Filter of Files for Reliable Non-Volatile Memory Systems

Cited by 5 publications

References 11 publications

Non-Volatile Memory File Systems: A Survey

Non-Volatile Memory File Systems: A Survey

A Real-Time Dependable Flash Storage System

Self-Learning Hot Data Prediction: Where Echo State Network Meets NAND Flash Memories

Contact Info

Product

Resources

About