Mohammad Arshad Email & Phone Number

Bengaluru, Karnataka, India

Key member of the engineering team that developed in-house Open-source Hadoop distribution and successfully replaced Hortonworks Data Platform (HDP).Worked on Ambari, Ambari Metrics, ZooKeeper, Sqoop, Bigtop etc. Handled backward incompatibility issues, fixed bugs, enhanced, remediated CVEs, supported in cluster upgrade. Delivered some of these works with assistance of a junior team member.Stabilized and enhanced Apache Ambari metrics for 2000 node Hadoop cluster. Contributed some of the Apache Ambari work to open source, became Apache Ambari committer.

https://zookeeper.apache.org/credits.html

Bengaluru, Karnataka, India

1. Improved ZooKeeper read write performance by 18% compared to the base. Performed CPU, memory and lock profiling to find the improvements.2. Improved HBase read write performance by using better caching, reducing GC time, using better compression algorithm, using better data structure etc.

Lead Zookeeper team of 4 members. Upgraded ZooKeeper version to 3.5.6, added/re-implemented private features, supported integration and contributed few feature in open source.ContributedLead the work to automate hbase cluster repair through HBCK2 command to make maintenance personals job easy.Added new commands in HBCK2 to list and fix hbase cluster problems. Fixed existing hbck2 issues and contributed to open source.Took ownership of RSGroup feature, verified its functionality, fixed existing RSGroup issues and contributed to open source.

Bengaluru Area, India

Core member of team that designed and implemented new architecture of hbase secondary index.Initial design code is available at https://github.com/Huawei-Hadoop/hindex)Implemented secondary index handling in following features - Region split: rebuild the index after split- Snapshot: include index information in snapshot, create index when restored or cloned- Compaction: delete index data when user data deleted, handle TTL, compaction and direct delete- Replication: don’t replicate index data- Bulk load: before loading HFile into cluster if region splits, rebuild the indexes- WAL Player: recreate the index region boundaries are changedKey member of Data Lake Center (based on HDFS and HDFS Router) TRP team.Implemented features and open sourced few HDFS-13443, HDFS-13495

Bengaluru Area, India

MRS (similar to Amazon EMR):As a supporting member worked few months on Huawei enterprise-level big data clusters on the cloud MRS. Worked on Manager, broker, API-Gateway and console etc. Done coding in go lang.ZooKeeper:Took end to end ownership of ZooKeeper. Analyzed, designed and implemented ZooKeeper requirements. Fixed issues, supported customers, handled CIHCC Demo:Lead a team of 3 members to develop Huawei BigData capability showcasing application which was presented in Huawei Connect Conference. Worked as technical lead, customer interface person, individual contributor.

Bengaluru Area, India

Bengaluru Area, India

Bengaluru Area, India

Worked on Workflow management system- Workflow Engine - Workflow Admin Tool- Workflow Designer

Bengaluru Area, India

Worked on Workflow management system

Bachelor Of Engineering - Be, Computer Science

Intermediate, Physics, Chemistry, Mathematics

Class 6Th To High School, Science

I was the second topper of the school in High School board exams.

Nursery To 5Th Class