ARTÍCULO
TITULO

JAMPI: Efficient Matrix Multiplication in Spark Using Barrier Execution Mode

Tamas Foldi    
Chris von Csefalvay and Nicolas A. Perez    

Resumen

The new barrier mode in Apache Spark allows for embedding distributed deep learning training as a Spark stage to simplify the distributed training workflow. In Spark, a task in a stage does not depend on any other tasks in the same stage, and hence it can be scheduled independently. However, several algorithms require more sophisticated inter-task communications, similar to the MPI paradigm. By combining distributed message passing (using asynchronous network IO), OpenJDK?s new auto-vectorization and Spark?s barrier execution mode, we can add non-map/reduce-based algorithms, such as Cannon?s distributed matrix multiplication to Spark. We document an efficient distributed matrix multiplication using Cannon?s algorithm, which significantly improves on the performance of the existing MLlib implementation. Used within a barrier task, the algorithm described herein results in an up to 24% performance increase on a 10,000 × 10,000 square matrix with a significantly lower memory footprint. Applications of efficient matrix multiplication include, among others, accelerating the training and implementation of deep convolutional neural network-based workloads, and thus such efficient algorithms can play a ground-breaking role in the faster and more efficient execution of even the most complicated machine learning tasks.

 Artículos similares

       
 
Leila Mosleh, Masoud Negahban-Azar and Mitchell Pavao-Zuckerman    
Urban areas are increasingly vulnerable to the effects of climate change. Stormwater Green infrastructure (SWGI) is seen as an approach to increase the climate resilience of urban areas, because they can buffer precipitation changes brought on by climate... ver más
Revista: Water

 
Muzamil Hussain Syed, Tran Quoc Bao Huy and Sun-Tae Chung    
With the rapid growth of internet data, knowledge graphs (KGs) are considered as efficient form of knowledge representation that captures the semantics of web objects. In recent years, reasoning over KG for various artificial intelligence tasks have rece... ver más

 
Andressa Trentin, Victória Hellen Chagas, Mayara Carla Uvida, Sandra Helena Pulcinelli, Celso Valentim Santilli and Peter Hammer    
Organic-inorganic coatings based on polymethyl methacrylate (PMMA)?silica?lithium are an efficient alternative to protect metals against corrosion. Although the preparation methodology is established and the thin coatings (~10 µm) are highly protective, ... ver más

 
Ruyi Zhang, Songnian Li, Baojing Wei and Xu Zhou    
The division of the territorial space functional area is the primary method to study the rational exploitation and use of land space. The research on the Production?Living?Ecological Space (PLES) change and its motivating factors has major implications f... ver más

 
Zhenkai Yang, Yixin Hua, Yibing Cao, Xinke Zhao and Minjie Chen    
As a new product of the Internet and big data era, migration data are of great significance for the revealing of the complex dynamic network patterns of urban agglomerations and for studying the relations between cities by using the ?space of flows? mode... ver más