Making Sense of Relativistic Distributed Systems
- 1.3k Downloads
Linearizability, a widely-accepted correctness property for shared objects, is grounded in classical physics. Its definition assumes a total temporal order over invocation and response events, which is tantamount to assuming the existence of a global clock that determines the time of each event. By contrast, according to Einstein’s theory of relativity, there can be no global clock: time itself is relative. For example, given two events A and B, one observer may perceive A occurring before B, another may perceive B occurring before A, and yet another may perceive A and B occurring simultaneously,with respect to local time.
Here, we generalize linearizability for relativistic distributed systems using techniques that do not rely on a global clock. Our novel correctness property, called relativistic linearizability, is instead defined in terms of causality. However, in contrast to standard “causal consistency,” our interpretation defines relativistic linearizability in a manner that retains the important locality property of linearizability. That is, a collection of shared objects behaves in a relativistically linearizable way if and only if each object individually behaves in a relativistically linearizable way.
Unable to display preview. Download preview PDF.
- 1.Mars One, http://www.mars-one.com/
- 4.Bailis, P., Fekete, A., Ghodsi, A., Hellerstein, J.M., Stoica, I.: The potential dangers of causal consistency and an explicit solution. In: Proc. of the 5th ACM Symposium on Cloud Computing (2012)Google Scholar
- 5.Brewer, E.A.: Towards robust distributed systems (Invited Talk). In: Proc. of the 19th Symposium on Principles of Distributed Computing (2000)Google Scholar
- 6.Burckhardt, S., Gotsman, A., Yang, H.: Understanding eventual consistency. Microsoft Research Technical Report MSR-TR-2013-39 (2013)Google Scholar
- 7.De Candia, G., Hastorun, D., Jampani, M., Kakulapati, G., Lakshman, A., Pilchin, A., Sivasubramanian, S., Vosshall, P., Vogels, W.: Dynamo: Amazon’s highly available key-value store. In: Proc. of the 21st ACM Symposium on Operating Systems Principles (2007)Google Scholar
- 8.Einstein, A.: The Meaning of Relativity, 6th edn. Taylor & Francis (2003)Google Scholar
- 11.Golab, W., Higham, L., Woelfel, P.: Linearizable implementations do not suffice for randomized distributed computation. In: Proc. of the 43rd ACM Symposium on Theory of Computing, pp. 373–382 (2011)Google Scholar
- 12.Hay, J.Z.Y.: Using local time to reconstruct memory models for parallel computation. Final year project. National University of Singapore (2012)Google Scholar
- 13.Hay, J.Z.Y., Tay, Y.C.: Memory consistency for parallel systems: A reformulation without global time (keynote address). In: Proc. National Conference on High Performance Computing & Simulation (2013)Google Scholar
- 15.Jackson, J.: The interplanetary internet. IEEE Spectrum (2005)Google Scholar
- 19.Lloyd, W., Freedman, M.J., Kaminsky, M., Andersen, D.G.: Don’t settle for eventual: scalable causal consistency for wide-area storage with COPS. In: Proc. 23rd ACM Symposium on Operating Systems Principles, pp. 401–416 (2011)Google Scholar
- 21.Terry, D., Petersen, K., Spreitzer, M., Theimer, M.: The case for non-transparent replication: Examples from Bayou. IEEE Data Engineering Bull. 21(4), 12–20 (1998)Google Scholar