arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1606.08905cs.DC
收藏

knor:一个NUMA优化的内存、分布式和半外部存储器k-means库

knor: A NUMA-Optimized In-Memory, Distributed and Semi-External-Memory k-means Library

Disa Mhembere, Da Zheng, Carey E. Priebe, Joshua T. Vogelstein, Randal Burns

摘要:

k-means是最有影响力和最常用的机器学习算法之一。它的计算限制了许多统计分析和机器学习任务的性能和可扩展性。我们重新思考和优化现代NUMA架构的k均值,开发一种新的并行化方案,延迟和最大限度地减少同步障碍。\textit{k-means NUMA Optimized Routine}(\textsf{knor})库具有(i)内存中(\textsf{knori})、(ii)分布式内存(\textsf{knord})和(iii)半外部内存(\textsf{knors})模块,这些模块可以针对不同的内存和硬件预算从根本上提高k均值算法的性能。\textsf{knori}可将单机数据集的性能提高一个数量级或更多。\textsf{knors}提高了k均值算法在使用SSD的内存预算上的可扩展性。\textsf{knors}可在一台机器上扩展到数十亿个点,只使用分布式内存系统所需资源的一小部分。\textsf{knord}保留了\textsf{knori}的性能特征,同时通过云中的分布式计算在内存中进行扩展。\textsf{knor}修改埃尔坎的三角不等式修剪算法,使我们利用它对十亿点的数据集没有显着的内存开销的原始算法。我们证明了\textsf{knor}优于分布式商业产品,如H $_2$ O,Turi(以前的Dato,GraphLab)和Spark的MLlib超过一个数量级的数据集的$10^7$到$10^9$点。

英文摘要:

k-means is one of the most influential and utilized machine learning algorithms. Its computation limits the performance and scalability of many statistical analysis and machine learning tasks. We rethink and optimize k-means in terms of modern NUMA architectures to develop a novel parallelization scheme that delays and minimizes synchronization barriers. The \textit{k-means NUMA Optimized Routine} (\textsf{knor}) library has (i) in-memory (\textsf{knori}), (ii) distributed memory (\textsf{knord}), and (iii) semi-external memory (\textsf{knors}) modules that radically improve the performance of k-means for varying memory and hardware budgets. \textsf{knori} boosts performance for single machine datasets by an order of magnitude or more. \textsf{knors} improves the scalability of k-means on a memory budget using SSDs. \textsf{knors} scales to billions of points on a single machine, using a fraction of the resources that distributed in-memory systems require. \textsf{knord} retains \textsf{knori}'s performance characteristics, while scaling in-memory through distributed computation in the cloud. \textsf{knor} modifies Elkan's triangle inequality pruning algorithm such that we utilize it on billion-point datasets without the significant memory overhead of the original algorithm. We demonstrate \textsf{knor} outperforms distributed commercial products like H$_2$O, Turi (formerly Dato, GraphLab) and Spark's MLlib by more than an order of magnitude for datasets of $10^7$ to $10^9$ points.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑