> Markdown version of [/videos/626-in-memory-computing-the-big-picture](https://www.wearedevelopers.com/videos/626-in-memory-computing-the-big-picture). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# In-Memory Computing - The Big Picture

What if you could completely bypass traditional databases? Discover how in-memory computing eliminates object-relational mapping bottlenecks, achieving microsecond query latency and slashing cloud costs by 99%.

- **Speakers:** [Markus Kett](https://www.wearedevelopers.com/@markus-kett)
- **Event:** World Congress 2023
- **Published:** August 11, 2023
- **Duration:** 30:36
- **URL:** https://www.wearedevelopers.com/videos/626-in-memory-computing-the-big-picture

## Summary

Modern applications spanning AI, IoT, and high-performance computing demand speed, low storage costs, and simplistic implementation. However, developers continually hit latency bottlenecks driven by the object-relational impedance mismatch. Translating rich, complex object graphs from object-oriented languages into flat relational tables or NoSQL documents forces expensive runtime data conversion. As noted in the presentation, latency directly impacts revenue—a mere 100-millisecond delay can cost a business like Amazon 1% in sales, highlighting why traditional millisecond-level database queries must evolve to microsecond-level memory access.

To bridge this gap, the industry often layers complex multi-tier architectures using distributed caches, in-memory data grids, or dedicated in-memory databases. While these solutions reduce disk-read latency, they fail to resolve the core mapping bottleneck and introduce massive infrastructure overhead. Data must still be serialized, shuttled across network wires, and mapped via an ORM for every single read and write operation.

The ultimate breakthrough lies in "system prevalence" architecture, enabling database-less data processing. Using the open-source Eclipse Store engine, developers can maintain multi-modal object graphs directly in memory and query massive datasets in microseconds using tools like the Java Streams API. Instead of leaning on expensive database clusters, the application persists binary data directly to cloud blob storage, such as AWS S3. This eliminates data conversion overhead, bypasses traditional databases entirely, and flattens complex caching tiers, successfully reducing database-as-a-service cloud costs by up to 99%.

**Keywords:** in-memory computing, object-relational impedance mismatch, database latency reduction, system prevalence architecture, eclipse store, object graph persistence, distributed caching layers, in-memory data grids, cloud blob storage, java streams API querying, database-less data processing, ORM performance bottlenecks, AWS S3 database alternative, microsecond data access

## Chapters

1. **Critical factors driving modern application performance requirements** (00:00) — Modern applications demand high performance, low data storage costs, and simple technical implementations.
1. **Analyzing speed differences between memory and database queries** (05:37) — Executing code in memory provides extreme speed advantages compared to the high latency of database access.
1. **Understanding the object-relational impedance mismatch in traditional databases** (09:04) — Storing complex object graphs within relational database tables creates deep architectural incompatibilities.
1. **Evaluating NoSQL databases and data conversion bottlenecks** (12:07) — Despite specialized data structures, non-relational databases still impose heavy performance penalties through object generation overhead.
1. **Implementing distributed caches to reduce relational database load** (15:11) — Extracting data into local or clustered cache environments avoids severe disk IO penalties but adds system complexity.
1. **Comparing distributed caches with in-memory data grids** (18:09) — In-memory data grids build upon distributed cache architecture by allowing organizations to distribute computational algorithms entirely in memory.
1. **Identifying mapping overheads within in-memory database clusters** (20:30) — Because in-memory database nodes run on separate architectures, their overall speed is still constrained by data conversion mappings.
1. **Eliminating database mapping via system prevalence and blob storage** (23:35) — Storing application object structures directly into scalable cloud blob services prevents classical storage latency entirely.
1. **Simplifying enterprise architecture with database-less data processing** (28:01) — Leveraging memory replication directly against binary stores removes both clustering and mapping overheads for significant cloud savings.

## Related Moments

- [Q&A on database vendor lock-in and alternative architectural choices](https://www.wearedevelopers.com/videos/746-swapping-low-latency-data-storage-under-high-load) (from "Swapping Low Latency Data Storage Under High Load")
- [Slashing IT infrastructure costs by adopting memory persistence](https://www.wearedevelopers.com/videos/405-build-ultra-fast-in-memory-database-apps-and-microservices-with-java) (from "Build ultra-fast In-Memory Database Apps and Microservices with Java ")
- [Modern application stacks and real-time data requirements](https://www.wearedevelopers.com/videos/806-leveraging-real-time-data-in-fsis) (from "Leveraging Real time data in FSIs")
- [Recognizing architectural drivers pushing event streaming system adoption](https://www.wearedevelopers.com/videos/538-event-messaging-and-streaming-with-apache-pulsar) (from "Event Messaging and Streaming with Apache Pulsar")
- [The object-relational impedance mismatch in standard database programming](https://www.wearedevelopers.com/videos/405-build-ultra-fast-in-memory-database-apps-and-microservices-with-java) (from "Build ultra-fast In-Memory Database Apps and Microservices with Java ")
- [Accelerating data processing via resilient in-memory computing strategies](https://www.wearedevelopers.com/videos/405-build-ultra-fast-in-memory-database-apps-and-microservices-with-java) (from "Build ultra-fast In-Memory Database Apps and Microservices with Java ")

## Related Articles

- [Making Data Warehouses Fast: A Developer’s Story](https://www.wearedevelopers.com/magazine/107-making-data-warehouses-fast-a-developer-s-story)
- [What does the history of data storage tell us about the future?](https://www.wearedevelopers.com/magazine/495-what-does-the-history-of-data-storage-tell-us-about-the-future)
- [Why Event-Driven Architecture Isn’t About Speed (and When You Actually Need It)](https://www.wearedevelopers.com/magazine/745-why-event-driven-architecture-isn-t-about-speed-and-when-you-actually-need-it)
- [Introducing Redis Agent Memory Server](https://www.wearedevelopers.com/magazine/699-introducing-redis-agent-memory-server)

## Related Jobs

- [Principal Engineer - AI Search & Vector Infrastructure](https://www.wearedevelopers.com/jobs/ext/353953-principal-engineer-ai-search-vector-infrastructure) at **Redis**
- [Principal Engineer - AI Search & Vector Infrastructure](https://www.wearedevelopers.com/jobs/ext/319507-principal-engineer-ai-search-vector-infrastructure) at **Redis**
- [Principal Engineer - AI Search & Vector Infrastructure](https://www.wearedevelopers.com/jobs/ext/381484-principal-engineer-ai-search-vector-infrastructure) at **Redis**
- [Principal Software Engineer, Database Infrastructure](https://www.wearedevelopers.com/jobs/ext/1465908-principal-software-engineer-database-infrastructure) at **GitHub**
- [Senior Backend Engineer (Java)](https://www.wearedevelopers.com/jobs/ext/19369-senior-backend-engineer-java) at **Bonial International GmbH**
- [Data Engineer](https://www.wearedevelopers.com/jobs/ext/1276933-data-engineer) at **PROSOZ Herten GmbH**