FlashStore

Debnath, Biplob; Sengupta, Sudipta; Li, Jin

doi:10.14778/1920841.1921015

Cited by 166 publications

(17 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work is about implementing efficient set data structures for NVRAM. Sets (most notably hash maps) are widely used, e.g., for key-value storage [Debnath et al 2010;Nishtala et al 2013;Raju et al 2017]. It is, therefore, expected that durable sets would be of high importance when NVRAMs reach mass production.…”

Section: Introductionmentioning

confidence: 99%

Efficient lock-free durable sets

Zuriel

Friedman

Sheffi

et al. 2019

Proc. ACM Program. Lang.

View full text Add to dashboard Cite

Non-volatile memory is expected to co-exist or replace DRAM in upcoming architectures. Durable concurrent data structures for non-volatile memories are essential building blocks for constructing adequate software for use with these architectures. In this paper, we propose a new approach for durable concurrent sets and use this approach to build the most efficient durable hash tables available today. Evaluation shows a performance improvement factor of up to 3.3x over existing technology.

show abstract

Section: Introductionmentioning

confidence: 99%

Efficient lock-free durable sets

Zuriel

Friedman

Sheffi

et al. 2019

Proc. ACM Program. Lang.

View full text Add to dashboard Cite

show abstract

“…There are many other types of KV storage systems to improve the read performance 4,5,23,24 . FAWN 23 stores the KV pairs in a log file in SSD, which can only be appended, and the system maintains a hash table in memory to enhance the speed of data lookups.…”

Section: Background and Motivationmentioning

confidence: 99%

“…FAWN 23 stores the KV pairs in a log file in SSD, which can only be appended, and the system maintains a hash table in memory to enhance the speed of data lookups. FlashStore 5 and SkimpyStash 4 also adopt the idea of FAWN, but both perform some optimization on the hash table in memory. Specifically, FlashStore leverages the Cuckoo 29 hash for memory managements while SkimpyStash transfers part of index in memory to SSD.…”

Section: Background and Motivationmentioning

confidence: 99%

“…Key‐value (KV) store has become an important part of NoSQL databases, and made many important achievements in the field of scientific research and various application areas in recent years. For example, Google's BigTable 1 and LevelDB 2 are often used in web indexing, MongoDB 3 and SkimpyStash 4 in online gaming, FlashStore 5 in data deduplication. Other examples include Dynamo 6 Amazon uses to solve the large amount of redundant data produced in e‐commerce, LinkBench 7 and RocksDB 8 Facebook adopts to maintain social networking, and Atlas 9 Baidu leverages to store cloud data.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Improving LSM‐trie performance by parallel search

et al. 2020

View full text Add to dashboard Cite

Summary LSM‐trie‐based key‐value (KV) store is often used to manage an ultralarge dataset in reality by introducing a number of sublevels at each level, its linear growth pattern can fairly reduce the write amplification in store operations. Although this design is effective for the write operation, the last level holds a large proportion of KV items, leading to the extreme imbalance of data distribution. Therefore, to support efficient read, we need to carefully consider this imbalance. On the other hand, to ensure that acquired data is latest, the LSM‐trie needs to search the dataset at different levels one by one, and this search method may take a lot of unnecessary time. When the number of items is ultralarge, the random lookup performance may be poor due to the imbalance data distribution. To address this issue, we improve the read performance of the LSM‐trie by changing its serial search to parallel search, using two threads to simultaneously search at the last level and other levels, respectively. Our experiment results show that the read performance of the LSM‐trie can be improved up to 98.35% and on average 71.55%.

show abstract

“…For example, RIPQ [46] is a novel caching layer that minimizes write amplification when using local SSD as a read-only cache for remote photo storage. FlashStore [24] is a key-value store designed to use SSD as a fast cache between DRAM and HDD, similarly minimizing the number of reads/writes done to SSD. Nitro [33] is an SSD caching system that uses data deduplication and compression to increase capacity.…”

Section: Related Workmentioning

confidence: 99%

Strata

Kwon

Fingler

Hunt

et al. 2017

Proceedings of the 26th Symposium on Operating Systems Principles

122

View full text Add to dashboard Cite

Current hardware and application storage trends put immense pressure on the operating system's storage subsystem. On the hardware side, the market for storage devices has diversified to a multi-layer storage topology spanning multiple orders of magnitude in cost and performance. Above the file system, applications increasingly need to process small, random IO on vast data sets with low latency, high throughput, and simple crash consistency. File systems designed for a single storage layer cannot support all of these demands together. We present Strata, a cross-media file system that leverages the strengths of one storage media to compensate for weaknesses of another. In doing so, Strata provides performance, capacity, and a simple, synchronous IO model all at once, while having a simpler design than that of file systems constrained by a single storage device. At its heart, Strata uses a log-structured approach with a novel split of responsibilities among user mode, kernel, and storage layers that separates the concerns of scalable, high-performance persistence from storage layer management. We quantify the performance benefits of Strata using a 3-layer storage hierarchy of emulated NVM, a flash-based SSD, and a high-density HDD. Strata has 20-30% better latency and throughput, across several unmodified applications, compared to file systems purpose-built for each layer, while providing synchronous and unified access to the entire storage hierarchy. Finally, Strata achieves up to 2.8× better throughput than a block-based 2-layer cache provided by Linux's logical volume manager. CCS CONCEPTS • Information systems → Hierarchical storage management; Storage class memory; • Software and its engineering → File systems management;

show abstract

FlashStore

Cited by 166 publications

References 16 publications

Efficient lock-free durable sets

Efficient lock-free durable sets

Improving LSM‐trie performance by parallel search

Strata

Contact Info

Product

Resources

About