Organizations like Baidu and Barclays have deployed Alluxio with Spark in their architecture, and have achieved impressive benefits and gains. Recently, Qunar deployed Alluxio with Spark in production and found that Alluxio enables Spark streaming jobs to run 15x to 300x faster. In their blog post, they described how Alluxio improved their system architecture, and mentioned that some existing Spark jobs would slow down or would never finish because they would run out of memory.
Testing distributed systems at scale is typically a costly yet necessary process. At Alluxio we take testing very seriously as organizations across the world rely on our technology, therefore, a problem we want to solve is how to test at scale without breaking the bank. In this blog we are going to show how the maintainers of the Alluxio open source project build and test our system at scale cost-effectively using public cloud infrastructure. We test with the most popular frameworks, such as Spark and Hive, and pervasive storage systems, such as HDFS and S3. Using Amazon AWS EC2, we are able to test 1000+ worker clusters, at a cost of about $16 per hour.
The cloud is rapidly becoming ubiquitous, with continued adoption focused on the flexibility and cost benefits of a utility infrastructure model. Enterprises are increasingly taking a “data first” view of infra- structure, which demands a new way of thinking in a world in which data is stored and accessed from multiple locations and providers. Performance and interoperability challenges, however, can present obstacles to cloud adoption and complicate data management. Techniques such as the use of data silos, ETL processes and multiple data copies, which are commonly employed to accommodate cloud limitations, often tend to offset the expected benefits of cloud infrastructure. Alluxio offers a new way to enhance the benefits of cloud infra- structure without the performance limitations or interoperability challenges resulting from accessing disparate data sources in multiple, often remote, locations.
Alluxio is an open source software solution that connects analytics applications to heterogeneous data sources through a data orchestration layer that sits between compute and storage. It runs on commodity hardware, creating a shared data layer abstracting the files or objects in underlying persistent storage systems. Applications connect to Alluxio via a standard interface, accessing data from a single unified source.
This white paper discusses the data center challenges Alluxio addresses, the benefits provided, and an overview of how it works.
Alluxio was created because we saw a need for innovation at the data layer rising from the growing complexity of connecting multiple compute frameworks to an ever-expanding mix of storage systems and formats. Our approach uses a memory-centric architecture that abstracts files and objects in underlying persistent storage systems and provides a shared data access layer for compute applications.
Alluxio is not a persistent storage system. Instead, Alluxio serves as a data access layer, residing between any persistent storage system (such as Amazon S3, Microsoft Azure Object Store, Apache HDFS or OpenStack Swift) and computation frameworks (such as Apache Spark, Presto or Hadoop MapReduce). This whitepaper provides a technical overview of the Alluxio architecture and describes the data flow for common read and write scenarios.
Small (kilobyte-sized) objects are the bane of highly scalable cloud object stores. Larger (at least megabytesized) objects not only improve performance, but also result in orders of magnitude lower cost, due to the current operation-based pricing model of commodity cloud object stores. For example, in Amazon S3’s current pricing scheme, uploading 1GiB data by issuing … Continued
The world is entering the data revolution era. Along with the latest advancements of the Internet, Artificial Intelligence (AI), mobile devices, autonomous driving, and Internet of Things (IoT), the amount of data we are generating, collecting, storing, managing, and analyzing is growing exponentially. To store and process these data has exposed tremendous challenges and opportunities. … Continued
From our friends at MOMO The hadoop ecosystem makes many distributed system/algorithms easier to use and generally lowers the cost of operations. However, enterprises and vendors are never satisfied with that, so higher performance becomes the next issue. We considered several options to address our performance needs and focused our efforts on Alluxio, which improves performance … Continued
Enterprises are adopting big data technologies to analyze and derive insight from their growing volumes of structured and unstructured data. A familiar problem is the requirement to analyze data from multiple independent storage silos concurrently. In order to consolidate the data, large enterprises typically use custom solutions or build a data lake. These approaches present additional challenges and can be costly and time consuming.