Sciweavers

FCSC
2016
8 years 17 days ago
An operational happens-before memory model
Happens-before memory model (HMM) is used as the basis of Java memory model (JMM). Although HMM itself is simple, some complex axioms have to be introduced in JMM to prevent the ca...
Yang Zhang, Xinyu Feng
FCSC
2016
8 years 17 days ago
Scene text detection and recognition: recent advances and future trends
Text, as one of the most influential inventions of humanity, has played an important role in human life, so far from ancient times. The rich and precise information embodied in te...
Yingying Zhu, Cong Yao, Xiang Bai
FAST
2016
8 years 17 days ago
sRoute: Treating the Storage Stack Like a Network
In a data center, an IO from an application to distributed storage traverses not only the network, but also several software stages with diverse functionality. This set of ordered...
Ioan A. Stefanovici, Bianca Schroeder, Greg O'Shea...
FAST
2016
8 years 17 days ago
The Composite-file File System: Decoupling the One-to-One Mapping of Files and Metadata for Better Performance
Traditional file system optimizations typically retain the one-to-one mapping of logical files to their physical metadata representations. This rigid mapping results in missed opp...
Shuanglong Zhang, Helen Catanese, An-I Andy Wang
FAST
2016
8 years 17 days ago
Towards Accurate and Fast Evaluation of Multi-Stage Log-structured Designs
Multi-stage log-structured (MSLS) designs, such as LevelDB, RocksDB, HBase, and Cassandra, are a family of storage system designs that exploit the high sequential write speeds of ...
Hyeontaek Lim, David G. Andersen, Michael Kaminsky
FAST
2016
8 years 17 days ago
Uncovering Bugs in Distributed Storage Systems during Testing (Not in Production!)
Testing distributed systems is challenging due to multiple sources of nondeterminism. Conventional testing techniques, such as unit, integration and stress testing, are ineffectiv...
Pantazis Deligiannis, Matt McCutchen, Paul Thomson...
FAST
2016
8 years 17 days ago
Using Hints to Improve Inline Block-layer Deduplication
Block-layer data deduplication allows file systems and applications to reap the benefits of deduplication without requiring per-system or per-application modifications. However...
Sonam Mandal, Geoff Kuenning, Dongju Ok, Varun Sha...
FAST
2016
8 years 17 days ago
The Tail at Store: A Revelation from Millions of Hours of Disk and SSD Deployments
We study storage performance in over 450,000 disks and 4,000 SSDs over 87 days for an overall total of 857 million (disk) and 7 million (SSD) drive hours. We find that storage pe...
Mingzhe Hao, Gokul Soundararajan, Deepak R. Kencha...
FAST
2016
8 years 17 days ago
NOVA: A Log-structured File System for Hybrid Volatile/Non-volatile Main Memories
Fast non-volatile memories (NVMs) will soon appear on the processor memory bus alongside DRAM. The resulting hybrid memory systems will provide software with submicrosecond, high-...
Jian Xu, Steven Swanson
FAST
2016
8 years 17 days ago
WiscKey: Separating Keys from Values in SSD-conscious Storage
We present WiscKey, a persistent LSM-tree-based key-value store with a performance-oriented data layout that separates keys from values to minimize I/O amplification. The design ...
Lanyue Lu, Thanumalayan Sankaranarayana Pillai, An...