How T3Go’s high-performance data lake using Apache Hudi and Alluxio shortened the time for data ingestion into the lake by up to a factor of 2. Data analysts using Presto, Hudi, and Alluxio in conjunction to query data on the lake saw queries speed up by 10 times faster.
Data challenges in today’s disaggregated world
Today we see more enterprise architectures shifting to hybrid and multi-cloud environments. And while this shift allows for more flexibility and agility, it also means having to separate compute from storage, creating new challenges in how data needs to be managed and orchestrated across frameworks, clouds, and storage systems.
Data is not local to compute, leading to degraded workload performance.
The same data needs to be accessible to different, popular analytical & ML frameworks.
Running computation where data persists makes scaling extremely limited & expensive.
No self service data
To make data accessible to the users, complex ETL jobs are needed that copy data across different silos.
high Cost of management
Cloud storage egress costs continue to rise due to multiple data storage layers in the cloud.
unreliable s3 performance
Today’s object storage capabilities are not ready for interactive big data workloads.
High availability, storage system data management and disaster recovery is complex.
limiteD DATA security
No unified way to secure data across different clouds and storage systems.
The need for a new DATA ORCHESTRATION platform
To address these data challenges, enterprises are adopting a new platform: the data orchestration platform. A unified data orchestration platform simplifies your data’s cloud journey.
A data orchestration platform fundamentally enables separation of storage and compute. It brings speed and agility to big data and AI workloads and reduces costs by eliminating data duplication and enables users to move to newer storage solutions like object stores.
Alluxio – BIG Data orchestration fRAMEWORK for The cloud
Alluxio is a compute agnostic, storage agnostic and cloud agnostic solution for big data and machine learning applications.
Data is local to compute, giving you memory-speed access for your big data and AI/ML workloads
Data is accessible through one unified namespace, regardless of where it resides
Data is as elastic as compute so you can abstract and independently scale compute and storage
This blog explores an innovative platform with Presto as the computing engine and Alluxio as a data orchestration layer between Presto and S3 storage, to support online services with instantaneous response within the gaming industry. The preliminary results show that Presto with Alluxio outperforms S3 significantly in all cases.Alluxio with metadata caching shows up to 5.9x performance gain when handling large numbers of small files.
Alluxio has made available a range of cloud offerings and integrations with the latest Alluxio version 2.1. At the first Data Orchestration Summit at the Computer History Museum, the company also announced the strengthening of partnerships with Amazon AWS and Google Cloud.