CAMR: Coded Aggregated MapReduce
This addresses scalability issues in big data processing for applications like deep learning, though it is incremental as it builds on prior work.
The paper tackles the communication bottleneck in MapReduce-like systems for distributed algorithms, particularly in deep learning, by proposing a new scheme that matches state-of-the-art load reduction while keeping the number of jobs and data splits small.
Many big data algorithms executed on MapReduce-like systems have a shuffle phase that often dominates the overall job execution time. Recent work has demonstrated schemes where the communication load in the shuffle phase can be traded off for the computation load in the map phase. In this work, we focus on a class of distributed algorithms, broadly used in deep learning, where intermediate computations of the same task can be combined. Even though prior techniques reduce the communication load significantly, they require a number of jobs that grows exponentially in the system parameters. This limitation is crucial and may diminish the load gains as the algorithm scales. We propose a new scheme which achieves the same load as the state-of-the-art while ensuring that the number of jobs as well as the number of subfiles that the data set needs to be split into remain small.