Amamria Amina

Member of the Log Analytics team, specializing in log centralization and management. Its role is to provide ITG teams with real-time processed data and to provide them with visualization, analysis, and machine learning tools to facilitate proactive monitoring and supervision of applications.

Scope

• Design and sizing of resilient and high-availability architectures

• Configuration and installation of "turnkey" Elasticsearch clusters for business teams

• Study and implementation of mapping, sharding, and replication strategies adapted to business needs

• Support for clients in the validation and integration of Logstash transformation pipelines

• Infrastructure monitoring

• Incident resolution and problem diagnosis

• Performance tuning to optimize search and indexing times

• Customer support in the data onboarding and exploitation process.

Implementations

• Implementation of CCR (Cross Cluster Replication) for the migration of 5 TB of data between two Elasticsearch clusters

• Design, sizing, and installation of several architectures centered on Elasticsearch and Kafka products

• Audit and performance tuning of a shared cluster to identify and eliminate bottlenecks and boost end-to-end ingestion times

• Implementation of a monitoring and alerting system based on Watchers

• Automation of Elasticsearch Log4j management through Ansible

• Study and major upgrade to version 8 of the Elastic Stack using Ansible

• Review and improvement of resilience and high availability aspects of an Elasticsearch cluster in the context of DR (Disaster Recovery)

Environment

Elastic Stack, Kafka, Git, Ansible, IBM Cloud, Python3, Linux Rhel, Hashicorp Vault, Opsgenie, Hashicorp Nomad, Jira, Confluence, Gitlab, Scrum

Member of the Data for Ops team, a union of Natixis-QOS and BPCE-ELK teams offering a log management service called Theia for Application Production teams and BPCE publishers. Theia aims to enable client teams to leverage their application and technical logs in real-time and optimize their service monitoring. Theia is based on Elastic Cloud Enterprise (ECE) and the Elastic Stack.

Scope

• Support business IT departments and application production management teams in using their data through dashboarding and machine learning

• Ensure daily follow-up with my manager, reporting sprint progress and encountered issues

• Provide support to business teams for sizing business clusters (on demand) and organizing their indexes/shards/replicas

• Maintain the ECE platform in operational and evolving condition

• Incident, change, and problem management

• Act as the primary point of contact for technical teams and teams consuming the Theia offering

• Study client business needs, develop, and prioritize corresponding stories

Implementations

• Introduction and implementation of the Agile Kanban methodology within the team

• Facilitation of daily stand-ups and agile rituals

• Design of the Theia architecture and writing of a complete technical document detailing the inventory of machines, hardware resources, data center, roles, bandwidth, network flows, listening ports…)

• Study of hardware and software publisher prerequisites to prepare the ECE platform for a daily indexing of 10 Terabytes.

• Preparation of an Ansible playbook to automate the installation of ECE (45 hosts) using Ansible Tower.

• Securing the platform with SSL and Kibana access through SAML.

• Creation of a deployment template compliant with a Hot-warm Cold-Frozen data tier architecture

Data consultant at Metrology As A Service, a cross-functional Cloud platform within the GTS (Global Technical Service) department. MAAS collects GTS data (API logs, metrics, repositories), processes it, and stores it in Elasticsearch (short retention) and Hadoop (long retention) for API monitoring through analysis and dashboarding. The MAAS service is partially "API-fied": clients can create Kafka authorizations and generate the creation of a standardized NiFi workflow (Kafka, Elasticsearch, Hadoop) with a single click.

Scope

• Implement, maintain, and evolve specific ingestion workflows in NiFi (collection, processing, and indexing)

• Prepare indexing templates in Elasticsearch

• Manage Elasticsearch version upgrades.

• Maintain the operational condition of the infrastructure: Kafka, NiFi, Elasticsearch, Hadoop

• Support GTS teams in exploiting their logs and metrics (dashboards, alerts, etc.)

Implementations

• Development of a Data Quality tool in Hadoop (duplicates, gaps) with PySpark.

• Upgrade of the Elasticsearch stack to version 6

• Development of a "Kafka Authorization Management" feature and its integration into the MAAS API.