Constructing the truly modern data application, capable of driving data-informed decisions across the organization, requires a lot of effort. To dramatically improve productivity, operational efficiency, and innovation, enterprises must acknowledge and deliver on modern data apps’ various requirements—including high concurrency, integration, and governance.
DBTA’s webinar, Succeeding with Modern Data Applications, gathered a series of experts to evaluate key technologies, tools, and best practices for developing modern data apps that enable a data-driven, innovation-forward enterprise.
Tyler Mitchell, senior product marketing manager, Couchbase, pointed out that there are a variety of business challenges facing today’s organizations—which are further complicated by the inadequacies of legacy architectures.
The AI architecture landscape introduces a plethora of challenges and opportunities. While the swath of moving parts makes it difficult—from data ingestion and preprocessing to autonomous agents—there are a variety of proven techniques that assist in driving both a flexible, yet controlled, AI implementation. These include:
- Unified data platform
- RAG, guardrails, keyword filters
- Automatic vectorization and indexing
- AI models running within your environment
- In-memory architecture
- Advanced semantic caching
- Short- and long-term memory for AI agents
- Transcripts of LLM interactions
Couchbase’s platform delivers a plethora of these techniques to enable the creation of modern data apps, including:
- Multipurpose access services: Transactional, analytical, mobile, and AI workloads
- In-memory architecture: Speed, caching, workload isolation, and auto-sharding
- Geo-distributed high availability: Unified management, auto-replicate across regions, cloud to edge
- Secure, flexible deployment: DBaaS, Kubernetes, on-premises, mobile
- Deployment tools and SDKs: Build faster, SQL++ JSON queries, AI frameworks, 12-plus SDK languages
Modern data applications are needed now more than ever, said Ripudaman Singh Kushwah, principal product manager, Informatica, due to the following business requirements:
- Decision velocity: Users need insights in seconds, not hours
- User expectations: Consumer-grade experiences in enterprise applications
- Competitive differentiation: Data apps are becoming the new battleground
- Scale requirements: Supporting millions of concurrent users and real-time processing
Yet, the data challenges exaggerated by AI—including traditional challenges such as data volume and velocity, as well as AI-amplified challenges such as vector data management and prompt engineering data—complicate architectural plans.
With Informatica’s Intelligent Data Management Cloud (IDMC) solution framework, enterprises can achieve a truly modern architecture that supports new evolutions for data applications. Benefits include:
- Unified Integration Platform: Cloud Application Integration (CAI) and Cloud Data Integration (CDI)
- Real-time processing: Advanced streaming and event-driven architecture
- AI-ready data preparation: Automated feature engineering and data quality
- Governance and Security: AI-aware cataloging and compliance
Which is paired with an iPaaS advantage:
- Pre-built connectors for 1000s of data sources and AI services
- Visual low-code/no-code development environment, reducing time to market
- Enterprise-grade security and compliance built in
Stephane Castellani, SVP marketing, CrateDB, defined what a modern application is: not just a dashboard, but an enabler of faster decision making, better customer experiences, and new revenue streams. Echoing previous speakers, Castellani asserted that addressing performance and scalability, concurrency, data governance, and integration challenges necessitates the examination of the following:
- Architecture choices: Adopt databases and platforms designed for real-time analytics, scalability, and mixed workloads (not just transactional or batch).
- Flexibility: Support structured, semi-structured, and unstructured data in one place.
- Developer enablement: Use familiar languages (such as SQL) to accelerate app development.
- Resilience and adaptability: Systems should adjust automatically to new business questions without constant re-optimization.
- Governance baked in: Security, lineage, and accuracy as part of the foundation, not an afterthought.
- AI integration: Feed AI/ML models with real-time features to go beyond insights and offer recommended actions.
Castellani then introduced CrateDB, a distributed SQL database offering performance at scale, designed for hyper-fast ingestion and sub-second queries over massive volumes, even billions of rows. This is ideal for supporting modern data apps, enabling real-time ingestion and querying; support for multiple data types; real-time aggregations, ad-hoc queries, search and AI integration; unified in a developer-friendly, adaptable, cloud native, distributed, and resilient platform.
Kevin Kline, technical evangelist, SolarWinds, emphasized the importance of observability and monitoring as part of developing and supporting modern data apps. The SolarWinds platform allows enterprises to evolve from monitoring to observability to fully autonomous operations, all at the organization’s unique pace. It offers:
- Automation and intelligence, driven by uniting granular, accurate, and trusted data and proving actionable insights
- Powerful, easy-to-use solutions deliver an integrated experience designed to make IT look easy by combining data from across your entire environment
- In-depth cross-domain analysis across the delivery chain and unparalleled architectural security power observability and service desk for the entire technology stack
- Designed to seamlessly connect with your critical business services via APIs, providing flexibility, visibility, and control—wherever your environment lives and wherever you’re going next
At its core, SolarWinds helps eliminate database performance problems to best enable modern data app success. It acknowledges that finding the root cause of performance problems is difficult; troubleshooting problems with OS and virtual environments is time-consuming and frustrating; keeping database systems running smoothly is a constant headache; and that SQL query tuning can be tedious; and addresses each concern intelligently.
This is only a snippet of the full Succeeding with Modern Data Applications webinar. For the full webinar, featuring more detailed explanations, a roundtable Q&A, and more, you can view an archived version of the webinar here.