RankMap: A Platform-Aware Framework for Distributed Learning from Dense Datasets
This addresses the challenge of scalable distributed learning for large datasets, offering significant performance gains for applications like sparse recovery and power iteration, though it is incremental as it builds on existing factorization and scheduling techniques.
The paper tackles the problem of efficiently executing iterative learning algorithms on massive dense datasets by introducing RankMap, a platform-aware framework that factorizes data into lower-rank subspaces to create sparse representations, resulting in up to two orders of magnitude improvements in memory usage, execution speed, and bandwidth while maintaining accuracy.
This paper introduces RankMap, a platform-aware end-to-end framework for efficient execution of a broad class of iterative learning algorithms for massive and dense datasets. Our framework exploits data structure to factorize it into an ensemble of lower rank subspaces. The factorization creates sparse low-dimensional representations of the data, a property which is leveraged to devise effective mapping and scheduling of iterative learning algorithms on the distributed computing machines. We provide two APIs, one matrix-based and one graph-based, which facilitate automated adoption of the framework for performing several contemporary learning applications. To demonstrate the utility of RankMap, we solve sparse recovery and power iteration problems on various real-world datasets with up to 1.8 billion non-zeros. Our evaluations are performed on Amazon EC2 and IBM iDataPlex servers using up to 244 cores. The results demonstrate up to two orders of magnitude improvements in memory usage, execution speed, and bandwidth compared with the best reported prior work, while achieving the same level of learning accuracy.