CScale – A Programming Model for Scalable and Reliable Distributed Applications
Today’s connected world demands applications that are responsive, always available, and can service a large number of users. However, the task of writing such applications is daunting, even for experienced developers. We propose CScale, a programming model that attempts to simplify this task. The objective of CScale is to let programmers specify their application’s core logic declaratively without explicitly managing distribution. CScale applications have simple semantics that simplify reasoning about correctness and enable testing and debugging on the single machine. In turn, the CScale runtime manages all aspects of execution of a CScale application on large clusters, including deployment, state management (replication and data partitioning) and fault tolerance. CScale ensures high availability by using distributed wait-free data structures to manage state. CScale does impose some constraints on the kind of operations clients can perform. However, we find that many real-world web applications can be naturally expressed using CScale.
Unable to display preview. Download preview PDF.
- 1.Agrawal, P., Silberstein, A., Cooper, B.F., Srivastava, U., Ramakrishnan, R.: Asynchronous view maintenance for vlsd databases. In: SIGMOD 2009, Stanford InfoLab (June 2009)Google Scholar
- 2.Faleiro, J., Rajamani, S., Rajan, K., Ramalingam, G., Vaswani, K.: Generalized lattice agreement. In: Principles of Distributed Computing (PODC) (July 2012)Google Scholar
- 5.Shapiro, M., Preguiça, N., Baquero, C., Zawirski, M.: Convergent and commutative replicated data types. Bulletin of the European Association for Theoretical Computer Science (EATCS) (104), 67–88 (2011)Google Scholar
- 6.Zhuge, Y., Garcia-molina, H., Wiener, J.L.: The strobe algorithms for multi-source warehouse consistency. In: International Conference on Parallel and Distributed Information Systems, pp. 146–157 (1996)Google Scholar