April 2026 Summaries

Aerospike Voyager is a new desktop application that simplifies the process of working with databases by providing a unified experience for data exploration and production code integration. Available for macOS, Windows, and Linux, Voyager allows users to connect to a cluster with one click, explore data visually with a hierarchical data browser, and apply server-side filtering that seamlessly transitions into production code without requiring rewrites. This continuity is upheld through the use of the same Aerospike Expression Language (AEL) in both the application's UI and production code, ensuring that filters created during exploration can directly be used in applications via the new Aerospike Java and Python SDKs. Additionally, the application supports AI-native development by allowing AI agents to interact with the same database clusters using the embedded MCP server, facilitating seamless query generation without the need for translation. Voyager also provides functionality for safely managing large queries with options to cancel costly operations mid-flight and set query timeouts, enhancing development efficiency.

Aerospike has developed new Java and Python SDKs to address the challenges of using traditional database clients in modern development environments, particularly with the rise of AI coding assistants. These SDKs, designed from the ground up, feature idiomatic APIs that simplify common operations, allowing developers to focus on application logic without delving into complex configuration details typically handled by site reliability engineers and database administrators. The new SDKs introduce the Aerospike Expression Language, which simplifies writing filters and conditional expressions, and are structured for compatibility with AI coding assistants like GitHub Copilot, ensuring more accurate code generation. Alongside these SDKs, Aerospike is releasing Voyager, a tool to make their database more accessible within modern development workflows. The SDKs are currently in Preview, allowing developers to provide feedback while maintaining compatibility with existing clients.

A recent benchmark by benchANT compared Aerospike Enterprise 8.0.4 and Apache Cassandra 5.0.3, focusing on performance and resilience under conditions of sustained mixed workloads and node failures, rather than ideal peak performance scenarios. The results revealed that Aerospike outperformed Cassandra by maintaining higher throughput and significantly lower latency, particularly under stress conditions. While both systems remained operational during node failures, Aerospike demonstrated more consistent performance, with tightly bounded latency, whereas Cassandra exhibited greater variability and degradation, especially at higher operational loads. This benchmark underscores the importance of predictable performance in distributed databases, highlighting how Aerospike's stability and resilience make it better suited for real-time, mission-critical applications where consistent responsiveness is crucial.

Document modeling has become a standard approach in modern software architecture, with JSON-like structures often used due to their similarity to code objects. Aerospike, a high-performance database, has addressed the inefficiencies developers face when working with nested data structures by introducing path expressions in its 8.1.2 release. This feature allows for efficient, server-side querying and filtering of nested data, eliminating the need for client-side filtering and reducing network overhead. Developers can now create secondary indexes on deeply nested data, facilitating precise data retrieval without compromising performance. The improvements have led to reduced memory allocation, increased network efficiency, and a cleaner developer experience, as confirmed by performance tests and feedback from development partners. The update also includes enhancements such as tighter access control for set indexes, better management of storage reclamation processes, and support for AlmaLinux, further solidifying Aerospike's role in scalable, high-performance database solutions.

Aerospike has achieved its fourth Data Breakthrough Award with Aerospike Database 8 being named NoSQL Solution of the Year, chosen from over 3,500 global nominations. This recognition highlights the database's ability to provide strict serializability of ACID transactions at scale without sacrificing performance, a common tradeoff in the past. Aerospike Database 8 is designed for modern enterprises running AI-powered production systems that require real-time data processing, offering continuous access to fresh data and fast feedback loops crucial for AI applications. Over half of Aerospike's revenue now comes from such AI workloads, reflecting a growing demand for robust data infrastructure. The company has a history of success at these awards, having previously won titles such as Graph DBS Solution of the Year. Aerospike will showcase its technology at Google Cloud Next in 2026, demonstrating its capabilities in real-time performance with agentic AI and generative AI.

In modern system performance metrics, the focus has shifted from average latencies to percentile metrics like P95 and P99 to better capture the behavior of slower requests, but even these are insufficient for AI systems due to their complex execution chains. AI applications often involve numerous internal operations such as model invocations and database queries, leading to more pronounced fan-out effects where small performance variations can accumulate into significant user-visible delays. This results in a situation where the extreme tail of the latency distribution, beyond P99, becomes critical to user experience, as even rare slow events can frequently impact performance when numerous operations are involved. Consequently, engineers are now emphasizing the importance of controlling tail latency to ensure predictable performance, rather than solely optimizing for peak throughput or traditional percentile metrics. As AI systems continue to grow in complexity, they highlight the need for architectures that can maintain tightly bounded latency distributions to provide stable and consistent user experiences.

LangGraph's integration with Aerospike introduces a robust persistence layer for handling state and data in agentic systems, aiming to enhance responsiveness, stability, and simplicity at scale. This collaboration focuses on using Aerospike's capabilities to implement a three-set layout for checkpointing, ensuring efficient state storage, fast recovery, and predictable execution continuity. The integration includes reference implementations such as AerospikeSaver and AerospikeStore, with Docker-based setup instructions available in the public repository. AerospikeSaver employs a three-set structure to store execution state, metadata, and history separately, optimizing performance by minimizing read amplification and enabling constant-time checkpoint retrieval. Meanwhile, AerospikeStore provides a deterministic key-value layout that leverages Aerospike's primary-key model, allowing for fast, scoped data access according to LangGraph's hierarchical namespace. This design ensures that agent systems can maintain low latency and high throughput, while native TTL-based cleanup manages execution state efficiently without requiring additional cleanup processes, making Aerospike a fitting backend for LangGraph's memory-intensive applications.