Reference : On the Synchronization Bottleneck of OpenStack Swift-like Cloud Storage Systems
Scientific journals : Article
Engineering, computing & technology : Computer science
Security, Reliability and Trust
On the Synchronization Bottleneck of OpenStack Swift-like Cloud Storage Systems
Ruan, Mingkang mailto [Tsinghua University > School of Software]
Titcheu Chekam, Thierry mailto [University of Luxembourg > Interdisciplinary Centre for Security, Reliability and Trust (SNT) >]
Zhai, Ennan mailto [Yale University > Computer Science]
Li, Zhenhua mailto [Tsinghua University > School of Software]
Liu, Yao mailto [SUNY Binghamton > Computer Science]
E, Jinlong mailto [Tsinghua University > Computer Science and Technology]
Cui, Yong mailto [Network Institute, Beijing > Department of Computer Science and Technology]
Xu, Hong mailto [City University of Hong Kong > Computer Science]
IEEE Transactions on Parallel and Distributed Systems
IEEE Xplore
Yes (verified by ORBilu)
[en] Cloud computing;Delays;Electronic mail;Open source software;Protocols;Reliability;Synchronization;Cloud storage;OpenStack Swift;object synchronization;performance bottleneck
[en] As one type of the most popular cloud storage services, OpenStack Swift and its follow-up systems replicate each object across multiple storage nodes and leverage object sync protocols to achieve high reliability and eventual consistency. The performance of object sync protocols heavily relies on two key parameters: r (number of replicas for each object) and n (number of objects hosted by each storage node). In existing tutorials and demos, the configurations are usually r = 3 and n < 1000 by default, and the sync process seems to perform well. However, we discover in data-intensive scenarios, e.g., when r > 3 and n >> 1000, the sync process is significantly delayed and produces massive network overhead, referred to as the sync bottleneck problem. By reviewing the source code of OpenStack Swift, we find that its object sync protocol utilizes a fairly simple and network-intensive approach to check the consistency among replicas of objects. Hence in a sync round, the number of exchanged hash values per node is Theta(n x r). To tackle the problem, we propose a lightweight and practical object sync protocol, LightSync, which not only remarkably reduces the sync overhead, but also preserves high reliability and eventual consistency. LightSync derives this capability from three novel building blocks: 1) Hashing of Hashes, which aggregates all the h hash values of each data partition into a single but representative hash value with the Merkle tree; 2) Circular Hash Checking, which checks the consistency of different partition replicas by only sending the aggregated hash value to the clockwise neighbor; and 3) Failed Neighbor Handling, which properly detects and handles node failures with moderate overhead to effectively strengthen the robustness of LightSync. The design of LightSync offers provable guarantee on reducing the per-node network overhead from Theta(n x r) to Theta(n/h). Furthermore, we have implemented LightSync as an open-source patch and adopted it to OpenStack Swift, thus reducing the sync delay by up to 879x and the network overhead by up to 47.5x.

File(s) associated to this reference

Fulltext file(s):

Open access
lightsync-TPDS.pdfAuthor postprint1.36 MBView/Open

Bookmark and Share SFX Query

All documents in ORBilu are protected by a user license.