Schlegel Diagram and Optimizable Immediate Snapshot Protocol

11/20/2017
by   Susumu Nishimura, et al.
0

In the topological study of distributed systems, the immediate snapshot is the fundamental computation block for the topological characterization of wait-free solvable tasks. However, in reality, the immediate snapshot is not available as a native built-in operation on shared memory distributed systems. Borowsky and Gafni have proposed a wait-free multi-round protocol that implements the immediate snapshot using more primitive operations, namely the atomic reads and writes. In this paper, up to an appropriate reformulation on the original protocol by Borowsky and Gafni, we establish a tight link between each round of the protocol and a topological operation of subdivision using Schlegel diagram. Due to the fact shown by Kozlov that the standard chromatic subdivision is obtained by iterated subdivision using Schlegel diagram, the reformulated version is proven to compute the immediate snapshot in a topologically smoother way. We also show that the reformulated protocol is amenable to optimization: Since each round restricts the possible candidates of output to an iteratively smaller region of finer subdivision, each process executing the protocol can decide at an earlier round, beyond which the same final output is reached no matter how the remaining rounds are executed. This reduces the number of read and write operations involved in the overall execution of the protocol, relieving the bottleneck of access to shared memory.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
04/19/2023

Distributed Multi-writer Multi-reader Atomic Register with Optimistically Fast Read and Write

A distributed multi-writer multi-reader (MWMR) atomic register is an imp...
research
01/09/2018

Tight Bounds on the Round Complexity of the Distributed Maximum Coverage Problem

We study the maximum k-set coverage problem in the following distributed...
research
12/12/2017

A characterization of colorless anonymous t-resilient task computability

A task is a distributed problem for n processes, in which each process s...
research
03/06/2022

Relaxed Paxos: Quorum Intersection Revisited (Again)

Distributed consensus, the ability to reach agreement in the face of fai...
research
09/08/2017

Efficient Logging in Non-Volatile Memory by Exploiting Coherency Protocols

Non-volatile memory (NVM) technologies such as PCM, ReRAM and STT-RAM al...
research
10/08/2018

Optimal Memory-Anonymous Symmetric Deadlock-Free Mutual Exclusion

The notion of an anonymous shared memory (recently introduced in PODC 20...
research
05/25/2021

The Topology of Randomized Symmetry-Breaking Distributed Computing

Studying distributed computing through the lens of algebraic topology ha...

Please sign up or login with your details

Forgot password? Click here to reset