Hadoop Administrator

Role: Hadoop Operations Admin

Experience: 3+yrs

Notice period: Only 15-30days

Location: Remote

Package: 10LPA max

1. Support for 2 different categories, Hadoop Platform Support for Hadoop on Google Cloud Dataproc Cluster Performance Optimization & Hive DB Support on Google Cloud Dataproc
2. Administer Customers Hadoop file system and platform including directories, disk space and scheduled platform-wide processes.
3. The service includes performing administrative functions including: Autoscaling, sizing up or down the number of worker nodes needed on the cluster Setting.
4. Google Monitoring for creating dashboard and monitoring performance, Setting Alerting for various Dataproc services and mapping it to PagerDuty Review.
5. Identity and Access Management (IAM) usage security administration for standard services available on cluster.
6. Applicable Release Management using GITTroubleshooting operational issues related to Hadoop servicesData refresh to lower lifecycle environments.
7. Space administration on VM instance and Dataproc Master.
8. Worker Node : Monitoring Space usage on Google Cloud StorageWork with Google support team for any server issue or job failure issue and implement solutions provided.
9. Make parameter changes or add new patches as solution to any job failures
10. Regularly monitor performance of jobs and cluster.
11. Provide feedback to Customer, Work with Application team to transfer data processed by Dataproc.
12. Automate administrative process like restarting service or monitoring health with help of AnsibleWork with SRE team to create/rebuild Dataproc cluster using Terraform
13. Work on Disaster Recovery switchover activity
14. Integration Hadoop environment with other components like TDCH, BigQuery, Query Grid.
15. Working on the connections testing Installation and maintenance of GDC-COP reporting tool for gathering data related to cluster and jobs performance on Dataproc
16. clustersMonitoring - Monitoring Hadoop supported operational services for successful data processing operations.
17. Security Administration - Work with Loblaw, Cloud Platform Support team and map permission and roles to be given in Cloud IAM.
18. Share the Privileged accounts of Data Integration hub whenever requested by Loblaw Security teams.
19. Release Management Support Customer in promoting release of new platform objects (e.g. files, jars etc.), providing access rights from lower environment to higher environment.
20. Once a new process is released into higher environment, the Hadoop administrator will verify the deployed objects against pre-agreed operational checklist.
21. Application Release Management - Hadoop Administration team will assist in releasing scripts / changes to existing workloads from development / QA to production environment.
22. Provide support for creating hive objects (databases and tables) with appropriate authorization policies in production as a part of release management process.
23. Troubleshooting application issue related to Hive - Troubleshooting and investigation of failure of jobs/applications and queries running on Hive DB.
24. Adhoc Request related to data movement between different environments for Hive: Work on adhoc request on moving Hive tables and underlying data from one environment to different environment.
25. Creating various schemas Staging, Core, Work, Snow etc. with the help of Development team for upgrade project and work on data copy within the current and newly created environments.
26. Implement data sync strategy
27. Performance Management:

• Performance recommendation at Hive Level.
• Analyzing Data SkewAnalyzing file format Recommending appropriate compression technique.
• Cost based optimization parameter recommendation
• Recommending best practices for Hive
• Recommendin Statistics collection.
• Check Explain plans of bad jobs and recommending appropriate fix.
• Review various ETL/DIH,TDCH jobs time to time
• Analyze runtime and report to concerned teams if found performance degradation.

1. work on resolution Cluster Performance Optimization: Review the roles of cluster nodes (master, worker, and edge/gateway nodes)Review the deployment/placement of master and slave daemons, as well as client software, libraries, and gateways. Assess the choice and placement of back-end RDBMS systems, and their configurations.
2. Review the configurations of Dataproc (Hadoop) components.
3. Identify performance bottlenecks on Dataproc (Hadoop) cluster
4. Recommend deployment and/or configuration changes, and remediate the identified issues.
5. Establishing and collecting performance metrics data and identifying bottlenecks.
6. The common utilities that support the other Hadoop modules:

• Dataproc Operations Role
• Review Trending Ticket Handling.
• 100% SLA Ticket Fulfillment.
• PagerDutyReview Identity and Access Management (IAM) usage security administration for standard services available.
• Review Day in the Life for Operations Dataproc.
• Automation of Manual Tasks.
• Monitoring all services and infrastructure issues.
• Space Savings for Optimal System Performance.
• Dataproc Performance / Code Analysis
• Autosys Jobs Analysis.
• Review Long Running jobs and high Memory utilization jobs.
• Provide suggestions on environmental parameter setup for jobs for optimization.
• Review and provide results on Performance Recommendation.
• Handle P1/P2 issues.
• Handle all priority issues with Google Vendor cases and tasks with Loblaw teams.
• Setup project wise Dashboard in GCP montioring console.
• Configure alerts based on threshold.
• Setup pagerduty notification based on severity.
• Create/Update Google cases based on severity.
• Setup bridge call involving Google team and respective stakeholders.
• Tasks for CRQ created by client and PCO team.
• Take care of pre-requisites and backup of existing configurations.