Integrated Bounds for Disintegrated Storage

We point out a somewhat surprising similarity between non-authenticated Byzantine storage, non-systematic coded storage, and emulations of shared registers from smaller ones. A common characteristic in all of these is the inability of reads to safely return a value obtained in a single atomic access to shared storage. We collectively refer to such systems as disintegrated storage, and show integrated space lower bounds for asynchronous regular wait-free emulations in all of them. In a nutshell, if readers are invisible, then the storage cost of such systems is inherently exponential in the size of written values; otherwise, it is at least linear in the number of readers. Our bounds are asymptotically tight to known algorithms, and thus justify their high costs.

[1]  Jennifer L. Welch,et al.  One-write algorithms for multivalued regular and atomic registers , 2000, Acta Informatica.

[2]  Alexander Spiegelman,et al.  Space Complexity of Fault-Tolerant Register Emulations , 2017, PODC.

[3]  Leslie Lamport,et al.  On interprocess communication , 1986, Distributed Computing.

[4]  Stefano Tessaro,et al.  Optimal Resilience for Erasure-Coded Byzantine Distributed Storage , 2005, International Conference on Dependable Systems and Networks (DSN'06).

[5]  Ghassan O. Karame,et al.  PoWerStore: proofs of writing for efficient and robust storage , 2012, CCS.

[6]  Idit Keidar,et al.  Space Bounds for Reliable Storage: Fundamental Limits of Coding , 2016, PODC.

[7]  Michael K. Reiter,et al.  Byzantine quorum systems , 1997, STOC '97.

[8]  Zhiying Wang,et al.  On multi-version coding for distributed storage , 2014, 2014 52nd Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[9]  Michael K. Reiter,et al.  Efficient Byzantine-tolerant erasure-coded storage , 2004, International Conference on Dependable Systems and Networks, 2004.

[10]  Idit Keidar,et al.  Wait-free regular storage from Byzantine components , 2007, Inf. Process. Lett..

[11]  Rida A. Bazzi,et al.  Non-skipping Timestamps for Byzantine Data Storage Systems , 2004, DISC.

[12]  Idit Keidar,et al.  Byzantine disk paxos: optimal resilience with byzantine shared memory , 2004, PODC '04.

[13]  Dan Dobre,et al.  Erasure-Coded Byzantine Storage with Separate Metadata , 2014, OPODIS.

[14]  Yuanhao Wei Space Complexity of Implementing Large Shared Registers , 2018, ArXiv.

[15]  Nancy A. Lynch,et al.  Information-Theoretic Lower Bounds on the Storage Cost of Shared Memory Emulation , 2016, PODC.

[16]  Marcos K. Aguilera,et al.  On using network attached disks as shared memory , 2003, PODC '03.

[17]  Leslie Lamport,et al.  On interprocess communication , 1986, Distributed Computing.

[18]  Gary L. Peterson,et al.  Concurrent Reading While Writing , 1983, TOPL.

[19]  Michael Dahlin,et al.  Minimal Byzantine Storage , 2002, DISC.

[20]  Sam Toueg,et al.  Fault-tolerant wait-free shared objects , 1992, Proceedings., 33rd Annual Symposium on Foundations of Computer Science.

[21]  Rachid Guerraoui,et al.  Optimistic Erasure-Coded Distributed Storage , 2008, DISC.

[22]  Yuanhao Wei,et al.  Step Optimal Implementations of Large Single-Writer Registers , 2016, OPODIS.

[23]  Nancy A. Lynch,et al.  A coded shared atomic memory algorithm for message passing architectures , 2014, 2014 IEEE 13th International Symposium on Network Computing and Applications.

[24]  Marcos K. Aguilera,et al.  Using erasure codes efficiently for storage in a distributed system , 2005, 2005 International Conference on Dependable Systems and Networks (DSN'05).

[25]  Rachid Guerraoui,et al.  Amnesic Distributed Storage , 2007, DISC.

[26]  Dan Dobre,et al.  On the Time-Complexity of Robust and Amnesic Storage , 2008, OPODIS.