Ayush Saxena

Working as part of Hive-Iceberg team, mainly focusing on areas related to Apache Hive-4.x Integration with Apache Iceberg

Working on solving production bugs and performance issues for Apache Hive.
Contributing the same to Apache Hive opensource community

Active member of Apache Hadoop opensource community.
Contributing/Reviewing/Commiting code fixes and improvements.

Working on analysing and optimising Hive ETL performance on Cloud storages along with handling feature requests, improvements and bugs around Hive Runtime framework

Worked on upgrading Hadoop supported version for Hive and adding support for Aarch64/ARM CPU architectures

Worked on integrating Apache Ozone with Hive

Worked on several improvements in Apache Tez

Worked on design and implementation of GeoSpatial Data support in Hive.
Integrating use of standard ESRI GeoSpatial UDFs in Hive along with providing an ability to store and import the standard ESRI geospatial data into Hive.

Hive Replication provides a mechanism to replicate hive metadata & data between clusters.

Worked on several feature enhancements, performance improvements and bug fixes for Hive Replication.

Designed & Implemented use of HDFS snapshot based data copy for External Table Data.

Designed & Implemented the Optimised Bootstrap Solution for Planned and Unplanned failovers.

Optimised the Checkpointing flow of Hive-Replication & Implemented several In-Progress tracking… Show more

Hive Replication provides a mechanism to replicate hive metadata & data between clusters.

Worked on several feature enhancements, performance improvements and bug fixes for Hive Replication.

Designed & Implemented use of HDFS snapshot based data copy for External Table Data.

Designed & Implemented the Optimised Bootstrap Solution for Planned and Unplanned failovers.

Optimised the Checkpointing flow of Hive-Replication & Implemented several In-Progress tracking mechanisms (metrics/JMX)

Implemented several data copy enhancements & solved some critical bugs.

Contributing the same to Apache Hive Opensource Show less

Worked on implementing several performance improvements & solving production issues for HWC & Spark-Acid.

Implemented a R interface for Spark-Hive Connectivity using sparklyr and HWC

Worked as part of Huawei Hadoop-HDFS team.
Contributed several performance improvements, bug fixes and solved some critical production issues.

Worked on features like HDFS Erasure Coding, Router Based Federation(RBF), ViewFs(Client Side Federation), WebHdfs, Block Storage Policies, Block Placement Policies(Developed Available Space Rack Fault Tolerant BPP), Observer Reads along with other core components of HDFS.

Contributed to Apache Hadoop

Worked on analysing and solving challenges on deploying Hadoop clusters on ARM machines.

Tool to move Data Replicas to different Disaster Recovery(DR) Availability Zones(AZ), depending upon the AZ policy set on the directory.

This enables to distribute the data replicas across different Availability Zones, in order to prevent data loss in case of failure of any Data Centre

Worked on providing an interface for the HDFS Java Client to interact with NAS Server similarly as it connects with a HDFS Namenode.

Worked as part of Huawei’s DataLake Client team in the later stages.
Chased some critical pre-release bugs and improvements. Worked on implementing the UserNamespace Server for DLC.
UNS provides a way to isolate FileSystem tree at user level

Added support for FLEX EC Native EC Algorithm(Similar to Intel ISA-L) in HDFS

Worked on stabilising and deployment of Hadoop-HDFS Erasure Coding feature, by contributing and reviewing bug fixes and performance improvements for the feature.

Feature owner for production issues and customer requests related to Erasure Coding at Huawei

Worked on stabilisation and deployment of Router Based Federation(RBF as a scalability solution for HDFS).
Contributed several critical bug fixes, code and performance improvements and Implemented several Client Protocol API's for RBF.