Accelerating Big Data Processing

Accelerating Big Data Processing and Associated Deep Learning on Modern Datacenters

A Tutorial to be presented at The 46th International Symposium on Computer Architecture (ISCA-2019)
by
Dhabaleswar K. (DK) Panda and Xiaoyi Lu (The Ohio State University)

When: 9:00am-12:30 pm, Sunday, June 23, 2019
Where: Room 104A, Phoenix Convention Center, Phoenix, Arizona, USA

Abstract

The convergence of HPC, Big Data, and Deep Learning is becoming the next game-changing business opportunity. Apache Hadoop, Spark, gRPC/TensorFlow, Kafka, and Memcached are becoming standard building blocks in handling Big Data oriented processing and mining. Modern HPC bare-metal systems and data center platforms have been fueled with the advances in multi-/many-core architectures, RDMA-enabled networking, NVRAMs, and NVMe-SSDs during the last decade. However, Big Data and Deep Learning middleware (such as Hadoop, Spark, Kafka, and gRPC) have not embraced such technologies fully. Recent studies have shown that default designs of these components can not efficiently leverage the features of modern HPC clusters, like Remote Direct Memory Access (RDMA) enabled high-performance interconnects, high-throughput parallel storage systems (e.g. Lustre), Non-Volatile Memory (NVM). In this tutorial, we will provide an in-depth overview of the architecture of Hadoop, Spark, Kafka, gRPC/TensorFlow, and Memcached. We will examine the challenges in re-designing networking and I/O components of these middleware with modern interconnects, protocols (such as InfiniBand, RoCE) and storage architectures. Using the publicly available software packages in the High-Performance Big Data project (HiBD, http://hibd.cse.ohio-state.edu), we will provide case studies of the new designs for several Hadoop/Spark/Kafka/gRPC/TensorFlow/Memcached components and their associated benefits. Through these, we will also examine the interplay between high-performance interconnects, storage (HDD, NVM, and SSD), and multi-core platforms (e.g., Xeon x86, OpenPOWER) to achieve the best solutions for these components and applications on modern HPC clusters and data center systems. We also present in-depth case-studies with modern Deep Learning tools (e.g., Caffe, TensorFlow, DL4J, BigDL) with RDMA-enabled Hadoop, Spark, and gRPC.

Targeted Audience and Scope

This tutorial is targeted for various categories of people working in the areas of Big Data and Deep Learning including high-performance Big Data and Deep Learning middleware, like Hadoop, Spark, Memcached, Kafka, gRPC, and TensorFlow, high performance communication and I/O architecture, storage, networking, cloud computing and data-intensive applications. Specific audience this tutorial is aimed at include:

Scientists, engineers, researchers, and students engaged in designing next-generation Big Data systems and applications
Designers and developers of Big Data and Deep Learning middleware, like Hadoop, Spark, Memcached, Kafka, gRPC, TensorFlow, etc.
Newcomers to the field of Big Data and Deep Learning who are interested in familiarizing themselves with Hadoop, Spark, Memcached, Kafka, TensorFlow, RDMA, and high-performance networking and I/O
Managers and administrators responsible for setting-up next generation Big Data and Deep Learning environments and high-end systems/facilities in their organizations/laboratories

The content level will be as follows: 30% beginner, 40% intermediate, and 30% advanced. There is no fixed pre-requisite. As long as the attendee has a general knowledge in Big Data, Deep Learning, Hadoop, Spark, Memcached, Kafka, TensorFlow, HPC, networking and storage architectures, and related issues, he/she will be able to understand and appreciate it. The tutorial is designed in such a way that an attendee gets exposed to the topics in a smooth and progressive manner. This tutorial is organized as a coherent talk to cover multiple topics.

Outline of the Tutorial

Introduction to Big Data Applications and Analytics
Overview of MapReduce and Resilient Distributed Datasets (RDD) Programming Models
Architecture Overview of Apache Hadoop, Spark, Kafka, gRPC, TensorFlow, and Memcached

MapReduce (V1 and YARN), HDFS, Spark, HBase
gRPC, TensorFlow
Kafka, Memcached

Overview of High-Performance Interconnects, Protocols, and Storage Architectures for Modern Datacenters

InfiniBand and RDMA
10/40 GigE, iWARP and RoCE technologies
SSD/NVM-based storage and Lustre parallel filesystem

Challenges in Accelerating Hadoop, Spark, gRPC/TensorFlow, Kafka, and Memcached on Modern Networking and Storage Architectures
Overview of Benchmarks and Applications using Hadoop, Spark, gRPC/TensorFlow, Kafka, and Memcached
Basic Acceleration Case Studies and In-Depth Performance Evaluation
- MapReduce over InfiniBand with RDMA, RAMDisk, SSD, and Lustre
- HDFS over InfiniBand with RDMA and Heterogeneous Storage (RAMDisk, SSD, HDD, and Lustre)
- Spark over InfiniBand with RDMA, SSD, and Lustre
- RPC over InfiniBand with RDMA
- Kafka over InfiniBand with RDMA and SSD
- Memcached over InfiniBand with RDMA, SSD, and Lustre
The High-Performance Big Data (HiBD) Project and Associated Releases
Other Activities for Accelerating Big Data Applications
Advanced Acceleration Case Studies

Hadoop over InfiniBand with RDMA and NVM
Erasure Coding for HDFS and Memcached
Deep Learning Tools (such as Caffe, TensorFlow, DL4J, BigDL) over RDMA-Hadoop and RDMA-Spark
Big Data over HPC Cloud

Conclusion and Q&A

Brief Biography of Speakers

Dr. Dhabaleswar K. (DK) Panda is a Professor of Computer Science at the Ohio State University. His research interests include parallel computer architecture, high performance computing, communication protocols, files systems, network-based computing, and Quality of Service. He has published over 450 papers in major journals and international conferences related to these research areas. Dr. Panda and his research group members have been doing extensive research on modern networking technologies including InfiniBand, HSE and RDMA over Converged Enhanced Ethernet (RoCE). His research group is currently collaborating with National Laboratories and leading InfiniBand and 10-100GigE/iWARP companies on designing various subsystems of next generation high-end systems. The MVAPICH2 (High Performance MPI over InfiniBand, iWARP and RoCE) open-source software libraries, developed by his research group, are currently being used by more than 3000 organizations worldwide (in 88 countries). This software has enabled several InfiniBand clusters (including the 3rd one) to get into the latest TOP500 ranking during the last decade. Dr. Panda's research is supported by funding from US National Science Foundation, US Department of Energy, and several industry including Intel, Cisco, SUN, Mellanox, QLogic, NVIDIA and NetApp. He is an IEEE Fellow and a member of ACM. More details about Dr. Panda are available at here.

Dr. Xiaoyi Lu is a Research Assistant Professor of the Department of Computer Science and Engineering at the Ohio State University, USA. His current research interests include high performance interconnects and protocols, Big Data, Hadoop/Spark/Memcached Ecosystem, Parallel Computing Models (MPI/PGAS), Deep Learning, Virtualization and Cloud Computing. He has published over 100 papers in International journals and conferences related to these research areas. Recently, Dr. Lu is leading the research and development of RDMA-based accelerations for Apache Hadoop, Spark, Kafka, HBase, and Memcached, and OSU HiBD micro-benchmarks, which are publicly available from http://hibd.cse.ohio- state.edu. These libraries are currently being used by more than 305 organizations from 35 countries. He is a core member of the MVAPICH2 project and he is leading the research and development of MVAPICH2-Virt (high-performance and scalable MPI for hypervisor and container based HPC cloud). He is a member of IEEE and ACM. More details about Dr. Lu are available at here.

Last Updated: June 21, 2019