Are You Picking the Right Database? – Summit 2025

Are You Picking the Right Database? – Summit 2025 | Presented by Datavail

Choosing the right database is critical to application performance, scalability, and long-term success. In this insightful session, Michael Agarwal, Director & Global Cloud Databases Practice Leader at Datavail, breaks down the key considerations for selecting the right database engine—whether relational, NoSQL, graph, time series, or in-memory.

This presentation explores the strengths, limitations, and ideal use cases for each type of database, helping organizations make informed decisions based on business needs, technical requirements, and modernization goals.

🔗 View the full presentation here: https://www.datavail.com/resources/are-you-picking-right-database/

🔍 Key Highlights
RDBMS vs. NoSQL: Understand the differences between relational databases (Oracle, SQL Server, PostgreSQL) and NoSQL options (MongoDB, Cassandra, DynamoDB).
Database Types Explained:
Document & Key-Value Stores: Ideal for flexible schemas and high-speed access.
Wide Column & Graph Databases: Perfect for distributed systems and relationship-heavy data.
Ledger & Time Series Databases: Designed for immutable logs and time-stamped data.
In-Memory Databases: Ultra-fast access for caching and session management.
Cloud Modernization: Learn how to migrate and modernize databases using AWS, Azure, and Oracle platforms.
6 R’s of Database Rationalization: Retain, Retire, Rehost, Replatform, Repurchase, Refactor.
Business Drivers for Modernization: Cost savings, scalability, performance, and reduced management overhead.
🎯 Why Watch This Session?
This session is ideal for:

IT leaders evaluating database platforms
Architects planning cloud migrations
Developers building scalable applications
DBAs managing diverse workloads
Anyone seeking to align database choices with business goals

1. Are You Picking the Right Database?

2. 25 years in IT Past: Software Developer, DBA, TPM, Cloud Architect, etc. Currently Global Practice Leader at Datavail (D atabase M igration / Modernizations , Site Reliability Engineering, and NoSQL practices) 8 years of experience with AWS & Azure Worked for Hertz & AWS in past www.datavail.com 2 M Michael Agarwal Datavail Director & Global Cloud Databases Practice Leader Connect or follow me on LinkedIn: linkedin.com/in/MichaelAgarwal/

3. Datavail at a Glance Delivering a superior approach to leveraging data through application of a tech-enabled global delivery model & deep specialization in databases, data management, and application services. $25 M Invested in IP that improves the service experience and drives efficiency 15 + Years building and operating mission critical data and application systems 1,000 + Employees www.datavail.com 3

4. Datavail is Growing Rapidly Because of Our Cloud Expertise Gold Partner 18+ years of experience 7 – Gold Competencies 2 – Silver Competencies 550+ Microsoft Consultant Certifications Recognized in 2017 by Microsoft as Cloud for Global Good Partner of the Year Advanced Tier Partner 9+ years experience 100+ AWS Certified Consultants Window Workload Competency One of 20 Database Specialist Partners 150+ successful AWS Migrations, Modernizations Service Delivery Partner – Amazon EC2, RDS, Aurora Oracle Partner 15+ years of experience 17 – Specializations 300+ Oracle Certifications Certified Cloud Excellence Implementor for EPM Strong relationship with Oracle’s EPM and Analytics product development org Regularly selected by Oracle to present at Industry Conferences

5. Fill Out the Evaluation Form For a Chance to Win a Bluetooth Speaker Compact design, vibrant sound quality at home and on the road Wireless, 33 ft range; 5W speaker Great for surround sound, music, audiobooks, podcasts, conference calls, zoom meetings, and more

6. Most common Oracle, SQL Server, PostgreSQL, MySQL, MariaDB, etc. You need referential integrity at the database engine level ACID (Atomicity, Consistency, Isolation, and Durability) transactions Horizontal scaling is limited to Read replica for the most part Examples: ERP, Banking transactions, Airline reservations, Retail Purchase but small / medium volume only RDBMS (Relational Database Management System)

7. Getting more popularity MongoDB, Cassandra, Couchbase, CouchDB, etc. AWS DynamoDB, Azure CosmosDB, AWS DocumentDB, AWS Keyspaces, Azure Managed Instance for Cassandra Horizontal scaling possible using Shards & Replication: Perfect match for high throughput / distributed applications Often settle for Eventual Consistency Multiple updates to same row: last one wins or conditional updates Referential Integrity can be applied through app layer NoSQL Databases

8. MongoDB: Most popular Document DB & Developers love it Cassandra: Read / Write every node, wide-column database AWS DynamoDB: Key Value, single digit millisecond response Azure CosmosDB: Swiss Army Knife, becoming “Azure Database Platform” since it supports many database types Examples: Online gaming scores, Real time bidding, Global scale applications etc. More on NoSQL Databases

9. Key – Value NoSQL Product ID Price Type Weight Color HP 1 $499 Appliance 60 lb 2 $10 Paint Purple 3 $100 Lawn Mower 10

10. Wide Column Databases Source: Scylladb.com Stored as {100001, fav color: Blue, fav shirt: golf shirt}

11. Less common but very useful for certain use cases Amazon Neptune, Neo4j, ArangoDB , Azure CosmosDB, etc. Uses graph structure: nodes and edges, edges define the relationship between the nodes (data) Relationship is a first-class citizen, so querying is not expensive Use cases: Social Media Networking, Medical research, Fraud Detection, Recommendation engine, etc. Graph Databases

12. More on Graph Databases Source: Neo4j.com

13. Ledger Database A type of NoSQL database that provides an immutable, transparent, and cryptographically verifiable transaction log that is owned by a central authority Meaning system-of-record (data integrity, completeness, and verifiability) If your application requires decentralization and involves multiple, untrusted parties, a blockchain solution may be appropriate Rare need Examples: Amazon QLDB Examples: Oracle, SQL Server, and many others are adding Ledger options in newer versions

14. Optimized for time-stamped or time series data Append only ingestion pattern Amazon TimeStream, InfluxDB , Prometheus, etc. Use cases such as IoT data, Application Monitoring, Observability, Vital signs, etc. Time Series Databases

15. Faster reads and writes from Cache (read in microseconds) Risk of data loss upon a process or server failure, but there are ways to mitigate that risk Often used in addition to another main database Redis, Memcached, MongoDB configured as in-memory compared to WiredTiger , etc. Example of use cases: Products Catalog, User Profiles, Session data, etc. In-Memory Databases

16. Database Migrations and Modernizations Amazon RDS Amazon EC2 Amazon Aurora Amazon DocumentDB Azure VM Amazon Keyspaces Azure Managed MySQL Azure Managed MariaDB Azure Managed SQL Azure Managed PostgreSQL Database Migration Service Database Migration Service Amazon DynamoDB Azure CosmosDB

17. Databases Portfolio Rationalization – 6 R’s www.datavail.com 17 Use Migration Tool Install Config Deploy Determine Platform Modify Infrastructure Buy Configure Redesign Code Changes Retain (Keep on Prem) Retire (Sunset App.) Rehost (IaaS) Replatform (PaaS) Repurchase (SaaS) Refactor (Build Cloud Native / open source / NoSQL) Determine Discover Production Transition Manual Automate Validation

18. Database Modernization Business Drivers www.datavail.com 18 Commercial licensed RDBMS  Open-Source RDBMS License cost savings RDBMS  NoSQL (DynamoDB / MongoDB / CosmosDB / Cassandra) Want horizontal scaling & distributed processing Instead of using RDBMS, consider using Cache DB, Graph DB, Elasticsearch, NoSQL, MPP etc. Using purpose-built databases RDS PostgreSQL  Aurora PostgreSQL PostgreSQL  Azure Hyperscale Citus Need better performance On-premise SQL Server  VM on Azure  Azure Managed Instance  Azure SQL DB SQL Server on EC2  AWS RDS SQL Server Less management overhead

19. Why Modernize to PaaS Databases? On-Premise Power / network / HVAC Hardware Lifecycle Server Maintenance Virtualization & OS Install OS Patching DB Software Install and Upgrades Database Backups HA, Scaling, DR Schema, Database code, Tuning IaaS Power / network / HVAC Hardware Lifecycle Server Maintenance Virtualization & OS Install OS Patching DB Software Install and Upgrades Database Backups HA, Scaling, DR Schema, Database code, Tuning Managed Service / PaaS Power / network / HVAC Hardware Lifecycle Server Maintenance Virtualization & OS Install OS Patching DB Software Install and Upgrades Database Backups HA, Scaling, DR Schema, Database code, Tuning

20. Cloud Modernization Legacy Databases Modern RDBMS & NoSQL Reliability problems Improved HA & DR Build / Release problems Security Gaps Highly Secure DevOps Monolithic Architecture Increasing Cost Cost Savings Microservices

21. Fill Out the Evaluation Form For a Chance to Win a Bluetooth Speaker Compact design, vibrant sound quality at home and on the road Wireless, 33 ft range; 5W speaker Great for surround sound, music, audiobooks, podcasts, conference calls, zoom meetings, and more

22. Thank You datavail.com [email protected] 866-811-6764