Learn about stream processing on Alluxio from real-world workloads at Qunar, as well as how to position Alluxio in the streaming architecture. Xueyan Li and Yupeng Fu explore how Alluxio has led to performance improvements averaging a 300x improvement at service peak time on stream processing workloads at Qunar.
China Unicom is one of the five largest telecom operators in the world. China Unicom’s booming business in 4G and 5G networks has to serve an exploding base of hundreds of millions of smartphone users. This unprecedented growth brought enormous challenges and new requirements to the data processing infrastructure. The previous generation of its data processing system was based on IBM midrange computers, Oracle databases, and EMC storage devices. This architecture could not scale to process the amounts of data generated by the rapidly expanding number of mobile users. Even after deploying Hadoop and Greenplum database, it was still difficult to cover critical business scenarios with their varying massive data processing requirements. The complicated the architecture of its incumbent computing platform created a lot of new challenges to effectively use resources.
Haoyuan Li explores Alluxio’s goal of making its product accessible to an even wider set of users, through a focus on security, new language bindings, and further increased stability. Haoyuan also covers some new APIs Alluxio is working on to allow applications to access data more efficiently and manage data across different under storage systems.
Using Alluxio, an open-source memory speed virtual distributed storage system, deployed on Mesos enables connecting any compute framework, such as Apache Spark, to storage systems via a unified namespace. Alluxio enables applications to interact with any data at memory speed. Alluxio can eliminate the pains of ETL and data duplication, and enable new workloads across all data. Adit will discuss the architecture of Mesos, Spark and Alluxio to achieve an optimal architecture for enterprises.
Alluxio has run in JD.com’s production environment on 100 nodes for six months. Mao Baolong, Yiran Wu, and Yupeng Fu explain how JD.com uses Alluxio to provide support for ad hoc and real-time stream computing, using Alluxio-compatible HDFS URLs and Alluxio as a pluggable optimization component. To give just one example, one framework, JDPresto, has seen a 10x performance improvement on average. This work has also extended Alluxio and enhanced the syncing between Alluxio and HDFS for consistency.
Cloud object storage systems provide different semantics and performance implications compared to HDFS. Applications like Presto cannot benefit from the node-level locality or cross-job caching when reading from the cloud. Deploying Alluxio with Presto to access cloud solves these problems because data will be retrieved and cached in Alluxio instead of the underlying cloud or object storage repeatedly. Bin will present the architecture to combine Presto with Alluxio with use cases from major internet companies like JD.com and NetEase.com, and their lessons learned to operate this architecture at scale.
Over the past two decades, the Big Data stack has reshaped and evolved quickly with numerous innovations driven by the rise of many different open source projects and communities. In this meetup, speakers from Uber, Alibaba, and Alluxio will share best practices for addressing the challenges and opportunities in the developing data architectures using new and emerging open source building blocks. Topics include data format (ORC) optimization, storage security (HDFS), data format (Parquet) layers, and unified data access (Alluxio) layers.
Presto is an open source distributed SQL engine widely recognized for its low-latency queries, high concurrency, and native ability to query multiple data sources. Alluxio is an open-source distributed file system that provides a unified data access layer at in-memory speed. The combination of Presto and Alluxio is getting more popular in many companies like JD, NetEase to leverage Alluxio as distributed caching tier on top of slow or remote storage for the hot data to query, avoiding reading data repeatedly from the cloud. In general, Presto doesn’t include a distributed caching tier and Alluxio enables caching of files and objects that the Presto query engine needs.
The data engineering team at Bazaarvoice, a software-as-a-service digital marketing company based in Austin, Texas, must handle data at massive Internet-scale to serve its customers. Facing challenges with scaling their storage capacity up and provisioning hardware, they turned to Alluxio’s tiered storage system and saw 10x acceleration of their Spark and Hive jobs running on AWS S3.
In this whitepaper you’ll learn:
Tags: apache hive, apache spark, aws s3, benchmark, case study, performance, tiered storage