Corpus ID: 396314

On the CALM Principle for Bulk Synchronous Parallel Computation

@article{Interlandi2014OnTC,
  title={On the CALM Principle for Bulk Synchronous Parallel Computation},
  author={Matteo Interlandi and L. Tanca},
  journal={ArXiv},
  year={2014},
  volume={abs/1405.7264}
}
In the recent years a lot of emphasis has been placed on two apparently disjoined fields: data-parallel and eventually consistent distributed systems. In this paper we propose a theoretical study over an eventually consistent data-parallel computational model. The keystone is provided by the recent finding that a class of programs exists which can be computed in an eventually consistent, coordination-free way: monotonic programs. This principle is called CALM and has been proven for distributed… Expand
On the CALM Principle for BSP Computation
TLDR
This paper sketches how, using novel techniques, the satisfiability of the if direction of the CALM principle can still be obtained, although just for a subclass of monotonic queries. Expand
A datalog-based computational model for coordination-free, data-parallel systems
TLDR
The case is made that the current form of CALM does not hold in general for data-parallel systems, and how, using novel techniques, the satisfiability of the CALM principle can still be obtained although just for the subclass of programs called connected monotonic queries. Expand

References

SHOWING 1-10 OF 69 REFERENCES
On the CALM Principle for BSP Computation
TLDR
This paper sketches how, using novel techniques, the satisfiability of the if direction of the CALM principle can still be obtained, although just for a subclass of monotonic queries. Expand
A new paradigm for parallel and distributed rule-processing
TLDR
This paper introduces a paradigm, called data-reduction, for the parallel evaluation of a general datalog program, and proves that decomposability, a concept introduced previously in [WS, CW], is undecidable. Expand
The declarative imperative: experiences and conjectures in distributed logic
TLDR
A case for database researchers to take a leadership role in addressing the impending programming crisis is made and a number of theoretical conjectures are presented that may both interest the database community, and clarify important practical issues in distributed computing. Expand
A bridging model for parallel computation
TLDR
The bulk-synchronous parallel (BSP) model is introduced as a candidate for this role, and results quantifying its efficiency both in implementing high-level language features and algorithms, as well as in being implemented in hardware. Expand
Win-move is coordination-free (sometimes)
TLDR
It is shown the surprising result that the query given by the well-founded semantics of the unstratifiable win-move program is coordination-free in some of the models the authors consider, and it is shown that the original transducer network model and variants form a strict hierarchy of classes of coordination- free queries. Expand
Giraph Unchained: Barrierless Asynchronous Parallel Execution in Pregel-like Graph Processing Systems
TLDR
The results demonstrate that the BAP model provides efficient and transparent asynchronous execution of algorithms that are programmed synchronously, and provides across-the-board performance improvements of up to 5× faster over synchronous systems and up to an order of magnitude faster than asynchronous systems. Expand
Parallelizing Datalog programs by generalized pivoting
TLDR
The concept of generalized pivoting is proposed, which is a sufficient condition for decomposability of arbitrary Datalog programs and the clam of completely decomposable programs is introduced and it is shown, that generalized pivototing is also a necessary condition for complete decomposality. Expand
Consistency Analysis in Bloom: a CALM and Collected Approach
TLDR
This work introduces Bloom, a distributed programming language that is amenable to high-level consistency analysis and encourages order-insensitive programming, and proposes a program analysis technique that identifies points of order in Bloom programs: code locations where programmers may need to inject coordination logic to ensure consistency. Expand
SYNC or ASYNC: time to fuse for distributed graph-parallel computation
TLDR
This paper proposes Hsync, a hybrid graph computation mode that adaptively switches a graph-parallel program between the two modes for optimal performance, and builds online sampling and offline profiling approaches combined with a set of heuristics to accurately predicting future performance in the two mode. Expand
Distributed processing of logic programs
TLDR
A new concept of predicate decomposability is defined, which means that the load of evaluating a predicate can be divided among a number of processors, without a need for communication among them, which results in a very significant speed-up of the evaluation process. Expand
...
1
2
3
4
5
...