Cosmos DB Backup, Migration and Indexing Assistance

@bindmission

Hello, I have 10 years of experience in Azure Cosmos DB management and optimization. I propose to assist with the backup of your 19GB Cosmos DB and the migration of a container locally to your laptop using AnyDesk. I will also create an index for RAG vector embeddings and optimize your queries for improved performance. My approach will ensure a seamless transition with no disruption to existing operations. Regards, VishnuLal NB

@Web1devloper

Hello, I have an extensive experience of 8+ years in DevOps, Cloud Computing. I have Deploy more than 100+ projects on VPS using Load Balancing and full automations with Modern Architecture . My Expertise in DevOps are: - • VPS (Linux, Windows, Mac) • Web Servers (Nginx, Apache) • Version Control Tools (Git, GitHub, GitLab) • CI/CD (GitHub Action, Jenkins) • Containerization Tools (Docker, Kubernetes) • System Administrator • Network Administrator • Cloud Service Provider ( AWS, Azure, GCP, Digital Ocean, Hostinger, GoDaddy) • DNS & Name Server Setup (SSL/TLS) • Infrastructure as Code (IaC) Tool (Terraform) • Configuration Management and Automation Tool (Ansible) • Security (SSL/TLS, Firewall) • Database (MySQL, MongoDB, Oracle, PostgreSQL, Firebase) • Load Balancing (Nginx, Apache) My Expertise in AWS are: - • EC2 (Elastic Ip's, Firewall, OS, Security Groups, Snapshots, Backup, Load Balancing, Cron Jobs) • ECS • Lambda Function • IAM Users • Amplify • S3 Bucket • Route53 • RDS (Relational Database Service) • Lightsail

@niveditaj1

Hello, I can perform this Cosmos DB optimization for you. Adding the vector embeddings to the indexing policy is absolutely the correct solution, and the performance gains will be significant, not marginal. Queries that perform vector searches without an index must scan the entire container, which explains the high RU cost and slow response time. My approach will be to first create the new container with an optimized indexing policy that includes your vector data. I will then perform a direct container-to-container migration. I will write a script to read the data in batches from the old container and write it to the new one using bulk operations to ensure the process is efficient. I will perform all of this work through the AnyDesk session as you require. 1) To confirm the workflow, you will log into the Azure portal via the AnyDesk session, and I will perform the configuration from there? 2) What is the property name for the vector embeddings in your JSON documents? 3) Instead of a local backup, is it acceptable to perform a direct container-to-container migration, which is faster and more reliable? Thanks, Nivedita

@AdityaNandwana19

Hi, I’ve worked with Azure Cosmos DB (Core SQL API) on projects where large datasets needed to be migrated, backed up, and restructured for performance. I can help you: Back up your 19GB Cosmos DB locally to your laptop. Create a new container with the right indexing policy (including your embeddings). Migrate your ~300k chunks safely into it without touching your live data. Optimise RU/s usage during import and after, so queries become faster and cheaper. Since you need this done over AnyDesk, I’ll handle the setup and scripts while you stay logged in to Azure, so nothing risky happens to your live database. I’ve done similar non-destructive migrations and indexing fixes before, and I can also give you a clear idea of whether adding embeddings to the index will bring real query speedups for your use case. Let’s connect, and I’ll walk you through the process step by step. Thanks, Aditya Nandwana