Brief Announcement: Efficient Recoverable Writable-CAS

被引:0
|
作者
Jayanti, Prasad [1 ]
Jayanti, Siddhartha [2 ,3 ]
Jayanti, Sucharita [4 ]
机构
[1] Dartmouth Coll, Hanover, NH 03755 USA
[2] Google Res, Mountain View, CA USA
[3] MIT, Cambridge, MA USA
[4] Brown Univ, Providence, RI 02912 USA
关键词
durable; recoverable; detectable; persistent memory; CAS;
D O I
10.1145/3583668.3594592
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
We present DuraCAS, a durable, i.e., recoverably linearizable and detectable implementation of the CAS (compare-and-swap) primitive. DuraCAS is writable, meaning it supports aWrite() operation along with CAS() and Read(); has constant time complexity per operation; allows for dynamic joining, meaning newly created processes (a.k.a. threads) of arbitrary names can join the protocol and access our implementation; and has adaptive space complexity, meaning the space use scales in the number of processes n that actually use the objects, as opposed to previous protocols whose space complexity depends on N, the maximum number of processes that the protocol is designed for. Furthermore, DuraCAS, requires only O(m + n) space to support m objects that get accessed by n processes, improving on the state-of-the-art O(m + n(2)). To our knowledge, DuraCAS is the first durable CAS algorithm that allows for dynamic joining, and is the first to exhibit adaptive space complexity.
引用
收藏
页码:366 / 369
页数:4
相关论文
共 50 条
  • [1] Brief Announcement: The Recoverable Consensus Hierarchy
    Golab, Wojciech
    PROCEEDINGS OF THE 2019 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING (PODC '19), 2019, : 212 - 214
  • [2] Brief Announcement: Benchmarking Recoverable Mutex Locks
    Xiao, Jeffrey
    Zhang, Zheng
    Golab, Wojciech
    PROCEEDINGS OF THE 32ND ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES (SPAA '20), 2020, : 583 - 585
  • [3] Brief Announcement: Detectable Sequential Specifications for Recoverable Shared Objects
    Li, Nan
    Golab, Wojciech
    PROCEEDINGS OF THE 2021 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING (PODC '21), 2021, : 557 - 560
  • [4] Brief Announcement: Consensus and Efficient Passive Replication
    Junqueira, Flavio
    Serafini, Marco
    DISTRIBUTED COMPUTING, DISC 2012, 2012, 7611 : 425 - 426
  • [5] Brief Announcement: On Solving Recoverable Mutual Exclusion Under System-Wide Failures
    Dhoked, Sahil
    Golab, Wojciech
    Mittal, Neeraj
    PROCEEDINGS OF THE 35TH ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES, SPAA 2023, 2023, : 287 - 290
  • [6] Brief Announcement: An Efficient Failure Detector for Omission Environments
    Cortinas, Roberto
    Soraluze, Iratxe
    Lafuente, Alberto
    Larrea, Mikel
    PODC 2010: PROCEEDINGS OF THE 2010 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING, 2010, : 83 - 84
  • [7] Brief Announcement: Communication Efficient Asynchronous Byzantine Agreement
    Patra, Arpita
    Rangan, C. Pandu
    PODC 2010: PROCEEDINGS OF THE 2010 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING, 2010, : 243 - 244
  • [8] Brief Announcement: Efficient Graph Algorithms without Synchronization
    Schneider, Johannes
    Wattenhofer, Roger
    PODC 2010: PROCEEDINGS OF THE 2010 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING, 2010, : 289 - 290
  • [9] Brief Announcement: Efficient Computation in Congested Anonymous Dynamic Networks
    Di Luna, Giuseppe A.
    Viglietta, Giovanni
    PROCEEDINGS OF THE 2023 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING, PODC 2023, 2023, : 176 - 179
  • [10] Brief Announcement: Concurrent Data Structures for Efficient Streaming Aggregation
    Cederman, Daniel
    Gulisano, Vincenzo
    Nikolakopoulos, Yiannis
    Papatriantafilou, Marina
    Tsigas, Philippas
    PROCEEDINGS OF THE 26TH ACM SYMPOSIUM ON PARALLELISM IN ALGORITHMS AND ARCHITECTURES (SPAA'14), 2014, : 76 - 78