A Transactional Model for Parallel Programming of Graph Applications on Computing Clusters

Abstract

We present here the results of our investigation of a transactional model of parallel programming on cluster computing systems. This model is specifically targeted for graph applications with the goal of harnessing unstructured parallelism inherently present in many such problems. In this model, tasks for vertex-centric computations are executed optimistically in parallel as serializable transactions. A key-value based globally shared object store is implemented in the main memory of the cluster nodes for storing the graph data. Task computations read and modify data in the distributed global store, without any explicitly programmed message-passing in the application code. Based on this model we developed a framework for parallel programming of graph applications on computing clusters. We present here the programming abstractions provided by this framework and its architecture. Using several graph problems we illustrate the simplicity of the abstractions provided by this model. These problems include graph coloring, k-nearest neighbors, and single-source shortest path computation. We also illustrate how incremental computations can be supported by this programming model. Using these problems we evaluate the transactional programming model and the mechanisms provided by this framework.

Incremental parallel computing for continuous queries in dynamic graphs using a transactional model

“…1 The task execution begins by first obtaining a start timestamp for the transaction. 14 This is a lock-free model of execution, and a transaction reads only the committed data.…”

“…An alternate approach is to execute the analytics program only initially and then perform incremental computations for supporting continuous queries as the graph data is modified. The primary focus of our work is on utilizing a transactional model of parallel computing for supporting continuous queries on dynamic graph structures using incremental computations. In this model, tasks for vertex‐centric computations are asynchronously executed in parallel as serializable transactions.…”

“…More details about the Beehive framework can be found in the work of Tripathi et al 1 Section 4 describes the approach to solve the five graph problems noted above using the transactional parallel programming model. The motivation in selecting these five problems is that we want to show how the incremental computation model performs on different types of graph problems.…”

Design of a Location-Based Publish/Subscribe Service Using a Graph-Based Computing Model

“…The Beehive [35] framework provides a transactional model of parallel computing on cluster computers for graph problems. Graph data is stored in a key-value based data store maintained in the RAM of the cluster computers.…”

“…In our approach, a location-based publish/subscribe service executes on a cluster of computers, which can be hosted on a cloud platform. In this investigation we use the Beehive [35] framework to show that its transactional model of parallel computing on graph data structures can be easily utilized in building location-based services hosted on a cluster computer. We present the design of a museum visitor service to illustrate our approach of using graph data structures at the core of its architecture, implemented using the graph data management facilities provided by the Beehive framework.…”