-
- Shiqin Yan
- University of Chicago, Chicago, IL
-
- Huaicheng Li
- University of Chicago, Chicago, IL
-
- Mingzhe Hao
- University of Chicago, Chicago, IL
-
- Michael Hao Tong
- University of Chicago, Chicago, IL
-
- Swaminathan Sundararaman
- Parallel Machines
-
- Andrew A. Chien
- University of Chicago, Chicago, IL
-
- Haryadi S. Gunawi
- University of Chicago, Chicago, IL
書誌事項
- タイトル別名
-
- Near-Perfect Elimination of Garbage Collection Tail Latencies in NAND SSDs
- 公開日
- 2017-08-31
- 権利情報
-
- https://www.acm.org/publications/policies/copyright_policy#Background
- DOI
-
- 10.1145/3121133
- 公開者
- Association for Computing Machinery (ACM)
この論文をさがす
説明
<jats:p> Flash storage has become the mainstream destination for storage users. However, SSDs do not always deliver the performance that users expect. The core culprit of flash performance instability is the well-known garbage collection (GC) process, which causes long delays as the SSD cannot serve (blocks) incoming I/Os, which then induces the long tail latency problem. We present <jats:sc>tt</jats:sc> F <jats:sc>lash</jats:sc> as a solution to this problem. <jats:sc>tt</jats:sc> F <jats:sc>lash</jats:sc> is a “tiny-tail” flash drive (SSD) that eliminates GC-induced tail latencies by circumventing GC-blocked I/Os with four novel strategies: plane-blocking GC, rotating GC, GC-tolerant read, and GC-tolerant flush. These four strategies leverage the timely combination of modern SSD internal technologies such as powerful controllers, parity-based redundancies, and capacitor-backed RAM. Our strategies are dependent on the use of intra-plane copyback operations. Through an extensive evaluation, we show that <jats:sc>tt</jats:sc> F <jats:sc>lash</jats:sc> comes significantly close to a “no-GC” scenario. Specifically, between the 99 and 99.99th percentiles, <jats:sc>tt</jats:sc> F <jats:sc>lash</jats:sc> is only 1.0 to 2.6× slower than the no-GC case, while a base approach suffers from 5–138× GC-induced slowdowns. </jats:p>
収録刊行物
-
- ACM Transactions on Storage
-
ACM Transactions on Storage 13 (3), 1-26, 2017-08-31
Association for Computing Machinery (ACM)
- Tweet
詳細情報 詳細情報について
-
- CRID
- 1361418521489309568
-
- DOI
- 10.1145/3121133
-
- ISSN
- 15533093
- 15533077
-
- データソース種別
-
- Crossref