ScyllaDB Challenges AWS with a 50% Cheaper DynamoDB Alternative

SUNNYVALE, Calif. – January 15, 2026 – ScyllaDB today announced the general availability of ScyllaDB X Cloud, a high-performance database-as-a-service that mounts a direct challenge to one of the cloud's most established services: Amazon Web Services' DynamoDB. The company is not only claiming superior performance and faster elastic scaling but is also making an aggressive financial guarantee: that its service will cost 50% less than DynamoDB for comparable workloads. This move targets the growing enterprise frustration with unpredictable cloud bills and the technical limitations of incumbent NoSQL databases.

Initially known as a high-performance alternative to Apache Cassandra, ScyllaDB has increasingly focused on the DynamoDB ecosystem. This new release marks its most significant push yet, aiming to lure away customers with a potent combination of cost savings, raw power, and strategic flexibility.

Challenging the Titan on Cost and Speed

The central claim of the ScyllaDB X Cloud launch is its promise to slash database costs in half compared to DynamoDB. This guarantee directly addresses a major pain point for users of AWS's popular NoSQL service: complex and often spiraling costs. DynamoDB's pricing model, which charges for provisioned or on-demand read and write capacity units, can force teams to overprovision resources to handle potential traffic spikes, leading to significant waste during off-peak hours.

ScyllaDB asserts its platform eliminates this need for overprovisioning. The company states ScyllaDB X Cloud can scale from 100,000 to 2 million operations per second in just minutes, all while maintaining consistent single-digit millisecond P99 latency. This rapid elasticity means capacity can be added precisely when needed and removed just as quickly, aligning costs much more closely with actual usage.

Making the transition feasible is ScyllaDB’s DynamoDB-compatible API, known as Alternator. This feature allows applications built for DynamoDB to run on ScyllaDB with minimal to no code changes. According to Todd Coleman, chief architect and technical co-founder at Yieldmo, this compatibility was a game-changer. “Its API was DynamoDB-compatible, meaning we could migrate with minimal code changes,” Coleman stated. “In fact, a single engineer implemented the necessary modifications in just a few days.”

Beyond cost and scaling, the new release also targets specific technical limitations of DynamoDB. Prem Kumar Patturaj, senior engineering manager at Freshworks, highlighted one such issue. “DynamoDB’s item size limitation forced multiple steps: we had to split the data into multiple records, write it, and then fetch it again,” he explained. “All of this logic ended up adding compute overhead. So, we migrated to ScyllaDB because these limitations don’t exist there.”

The Architecture of True Elasticity

ScyllaDB’s performance and cost claims are rooted in a 'close-to-the-metal' architecture designed to maximize the efficiency of modern hardware. The foundation is a shard-per-core design, where the database is partitioned so that each CPU core operates independently with its own dedicated memory and I/O, eliminating internal bottlenecks and enabling near-linear performance scaling.

The new ScyllaDB X Cloud builds on this with a 'tablet-based' system. The database is divided into small, fixed-size fragments of about 5 GB called tablets. Each tablet is a fully independent unit that can serve queries. When a cluster needs to scale up, new servers can begin accepting tablets and serving traffic almost immediately—often within ten seconds of joining the cluster. This fine-grained, rapid distribution of work is what enables the platform's hyper-elasticity.

This architectural approach yields several key benefits. It allows clusters to run at a much higher storage utilization, around 90% compared to the more typical 70%, reducing the number of required servers and lowering infrastructure costs. The platform also introduces storage autoscaling, where users define a minimum CPU footprint and storage scales automatically. Furthermore, the architecture’s efficiency is amplified on the latest cloud hardware. On AWS Graviton4-powered instances, ScyllaDB claims it can deliver up to double the throughput for the same price, showcasing how its software is optimized to extract maximum value from underlying infrastructure.

A Strategic Escape from Vendor Lock-In

While the technical merits are compelling, ScyllaDB X Cloud is also being positioned as a strategic tool for enterprises seeking to regain control over their cloud destiny. Dependency on proprietary services like DynamoDB can lead to vendor lock-in, making it difficult and expensive for companies to pursue multicloud or hybrid-cloud strategies.

By offering a DynamoDB-compatible API that can be deployed on any cloud or on-premises, ScyllaDB provides an explicit off-ramp from the AWS ecosystem. Companies can migrate existing applications with minimal friction, gaining the flexibility to deploy their database where it makes the most financial and operational sense. This resonates with a growing trend of organizations like Yieldmo, which cited the need for “cross-cloud deployments” as a key factor in their migration.

Other companies, such as ad-tech firm Digital Turbine, have made similar moves, migrating from DynamoDB to ScyllaDB to address performance and cost issues at scale. The ability to do so without a major application rewrite, thanks to the Alternator API, was a critical enabler. For many, this isn't just about reducing this month's bill; it's about long-term architectural freedom and mitigating the risk of being tied to a single provider's roadmap and pricing structure.

Navigating the Evolving NoSQL Battlefield

The Database-as-a-Service (DBaaS) market is a rapidly expanding and fiercely competitive space, with analysts projecting its global value to exceed $50 billion by 2030. While hyperscale providers like AWS, Microsoft, and Google offer dominant platforms such as DynamoDB, Azure Cosmos DB, and Google Cloud Firestore, a new wave of specialized vendors is carving out significant territory by addressing specific weaknesses of the incumbents.

ScyllaDB’s strategy is a prime example of this trend. Instead of competing on breadth of features, it focuses with laser precision on the needs of workloads that demand predictable, ultra-low latency at massive scale—and it does so with a clear, aggressive economic proposition. By directly targeting the pain points of a specific, widely used service, it offers a clear value proposition to a large, established user base.

This launch demonstrates a maturation of the cloud market, where one-size-fits-all solutions are increasingly being challenged by specialized alternatives that offer superior performance and cost-efficiency for demanding use cases. For enterprises running data-intensive applications, the availability of a viable, cost-effective, and flexible alternative to DynamoDB represents a significant shift in the balance of power, providing them with newfound leverage and control over their critical data infrastructure.