SBFT: a Scalable and Decentralized Trust Infrastructure

被引:201
作者
Gueta, Guy Golan [1 ]
Abraham, Ittai [1 ]
Grossman, Shelly [2 ]
Malkhi, Dahlia [1 ]
Pinkas, Benny [3 ]
Reiter, Michael [4 ]
Seredinschi, Dragos-Adrian [5 ]
Tamir, Orr [2 ]
Tomescu, Alin [6 ]
机构
[1] VMware Res, Palo Alto, CA 94304 USA
[2] Tel Aviv Univ, Tel Aviv, Israel
[3] BIU, Ramat Gan, Israel
[4] Univ N Carolina, Chapel Hill, NC USA
[5] Ecole Polytech Fed Lausanne, Lausanne, Switzerland
[6] MIT, Cambridge, MA 02139 USA
来源
2019 49TH ANNUAL IEEE/IFIP INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS (DSN 2019) | 2019年
关键词
DISTRIBUTED CONSENSUS; IMPOSSIBILITY;
D O I
10.1109/DSN.2019.00063
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
SBFT is a state of the art Byzantine fault tolerant state machine replication system that addresses the challenges of scalability, decentralization and global geo-replication. SBFT is optimized for decentralization and is experimentally evaluated on a deployment of more than 200 active replicas withstanding a malicious adversary controlling f = 64 replicas. Our experiments show how the different algorithmic ingredients of SBFT contribute to its performance and scalability. The results show that SBFT simultaneously provides almost 2x better throughput and about 1.5x better latency relative to a highly optimized system that implements the PBFT protocol. To achieve this performance improvement, SBFT uses a combination of four ingredients: using collectors and threshold signatures to reduce communication to linear, using an optimistic fast path, reducing client communication and utilizing redundant servers for the fast path. SBFT is the first system to implement a correct dual-mode view change protocol that allows to efficiently run either an optimistic fast path or a fallback slow path without incurring a view change to switch between modes.
引用
收藏
页码:568 / 580
页数:13
相关论文
共 72 条
  • [1] Abd-El-Malek Michael, 2005, ACM SIGOPS Operating Systems Review, V39, P59, DOI DOI 10.1145/1095810.1095817
  • [2] Abraham I., 2016, ABS161202916 CORR
  • [3] Abraham Ittai, 2018, ABS180110022 CORR
  • [4] Abraham Ittai, 2017, ABS171201367 CORR
  • [5] Prime: Byzantine Replication under Attack
    Amir, Yair
    Coan, Brian
    Kirsch, Jonathan
    Lane, John
    [J]. IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING, 2011, 8 (04) : 564 - 577
  • [6] Steward: Scaling Byzantine Fault-Tolerant Replication to Wide Area Networks
    Amir, Yair
    Danilov, Claudiu
    Dolev, Danny
    Kirsch, Jonathan
    Lane, John
    Nita-Rotaru, Cristina
    Olsen, Josh
    Zage, David
    [J]. IEEE TRANSACTIONS ON DEPENDABLE AND SECURE COMPUTING, 2010, 7 (01) : 80 - 93
  • [7] Scaling Byzantine fault-tolerant replication to wide area networks
    Amir, Yair
    Danilov, Claudiu
    Dolev, Danny
    Kirsch, Jonathan
    Lane, John
    Nita-Rotaru, Cristina
    Olsen, Josh
    Zage, David
    [J]. DSN 2006 INTERNATIONAL CONFERENCE ON DEPENDABLE SYSTEMS AND NETWORKS, PROCEEDINGS, 2006, : 105 - 114
  • [8] Hyperledger Fabric: A Distributed Operating System for Permissioned Blockchains
    Androulaki, Elli
    Barger, Artem
    Bortnikov, Vita
    Cachin, Christian
    Christidis, Konstantinos
    De Caro, Angelo
    Enyeart, David
    Ferris, Christopher
    Laventman, Gennady
    Manevich, Yacov
    Muralidharan, Srinivasan
    Murthy, Chet
    Binh Nguyen
    Sethi, Manish
    Singh, Gari
    Smith, Keith
    Sorniotti, Alessandro
    Stathakopoulou, Chrysoula
    Vukolic, Marko
    Cocco, Sharon Weed
    Yellick, Jason
    [J]. EUROSYS '18: PROCEEDINGS OF THE THIRTEENTH EUROSYS CONFERENCE, 2018,
  • [9] [Anonymous], 2017, IACR CRYPTOL EPRINT
  • [10] Aranha D.F., RELIC is an Efficient LIbrary for Cryptography