Ceph: a scalable, high-performance distributed file system

By S. Weil, Scott A. Brandt, Ethan L. Miller, Darrell D. E. Long, C. Maltzahn. USENIX Symposium on Operating Systems Design and Implementation, 2006

1,996 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/c5de8202da04a88a0122ac2c21780cd9ec3db43b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Tachyon: Reliable, Memory Speed Storage for Cluster Computing Frameworks (2014, background): “Distributed Storage Systems In big data analytics, distributed file systems [15, 45, 49], e.”
PNUTS: Yahoo!'s hosted data serving platform (2008, background): “Examples include Ceph [27], Boxwood [20], and Sinfonia [3].”
Biscuit: A Framework for Near-Data Processing of Big Data Workloads (2016, methodology): “In another example, typical Hadoop [8] and Ceph [39] deployments don’t use RAID.”
CRUSH: Controlled, Scalable, Decentralized Placement of Replicated Data (2006, methodology): “In our own performance tests of CRUSH as part of a distributed object-based storage system [Weil et al. 2006], we found that randomizing object placement resulted in an approximately 5% penalty in write performance due to variance in the OSD workloads, related in turn to the level of variation in…”
Secure data deduplication (2008, methodology): “Highperformance distributed file systems such as the Panasas Par llel File System [37], Ceph [35], and Lustre [7] typically have mu ch less security than “standard” distributed file systems, tra ding higher performance for lower security.”
f4: Facebook's Warm BLOB Storage System (2014, background): “Notable recent examples include the Google File System [18], BigTable [12], and Ceph [53].”
The Hadoop Distributed File System (2010, methodology): “Ceph [17] has a cluster of namespace servers (MDS) and uses a dynamic subtree partitioning algorithm in order to map the namespace tree to MDSs evenly.”
Fault Tolerance in Iterative-Convergent Machine Learning (2018, background): “Checkpoints are saved to shared persistent storage, such as distributed ﬁlesystems like NFS (Sandberg et al., 1988), CephFS (Weil et al., 2006), or distributed databases like Cassandra (Lakshman and Malik, 2010).”
Finding a Needle in Haystack: Facebook's Photo Storage (2010, background): “Haystack is an object store [7, 10, 12, 13, 25, 26] that we designed for sharing photos on Facebook where data is written once, read often, never modified, and rarely deleted.”
DepSky: dependable and secure storage in a cloud-of-clouds (2011, methodology): “This library offers an object store interface [Gibson 1998], similar to what is used by parallel ﬁle systems (e.g., [Ghe-mawat 2003, Weil 2006]), allowing reads and writes in the back-end (in this case, the untrusted clouds).”