ScyllaDB X Cloud update addresses database cost, performance

Prioritizing price and performance

Many enterprises have struggled with the cost of cloud computing dating back to the last decade. Economic uncertainty following the COVID-19 pandemic and rising inflation exacerbated concerns over cloud spending in the early 2020s. And over the past few years, expenses related to exploding interest in developing AI tools -- which require massive volumes of high-quality data to properly perform -- have proven prohibitive for some organizations.

The general availability of ScyllaDB X Cloud is a significant milestone for the company in delivering a new architecture that delivers enhanced elasticity and flexibility.

In response, some vendors have prioritized cost certainty and performance efficiency to help customers properly plan for their data management and AI development initiatives.

For example, cost control was at the core of numerous AWS platform updates unveiled during the tech giant's re:Invent conference in December 2025, including a new pricing model for its databases and performance improvements aimed at cutting costs in half. Similarly, database vendors Aerospike and Neo4j made performance the focus of recent releases.

Now, driven by customer feedback, according to co-founder and CEO Dor Laor, ScyllaDB's X Cloud update includes cost control measures and capabilities aimed at improving performance.

"[Cost control] was high on the list of customer feature requests," he said. "The key reason is that our customers really appreciated the price [and] performance of the previous generation, but its scaling speed was unpredictable. Adding nodes one at a time was particularly slow for, say, clusters that span five regions with three zones each."

Tablet-based autoscaling increases or decreases the resources needed to run workloads depending on demand, helping customers reduce wasteful spending on compute power that is not needed. Advanced data compression, meanwhile, dramatically lowers the amount of storage required to house a given amount of data, which also lowers expenses. Finally, improved storage utilization reduces the number of servers required to store data to lower infrastructure costs.

In addition, ScyllaDB now guarantees that the cost of using X Cloud is 50% or lower than the cost of using Amazon DynamoDB for the same storage and workloads.

Given that the X Cloud update emphasizes a managed approach to scaling database workloads and extends tablet-based elasticity to use cases including CDC, it is significant for ScyllaDB customers, according to Devin Pratt, an analyst at IDC.

"ScyllaDB X Cloud aims to make scaling feel routine, with predictable performance even as capacity changes," he said.

Beyond features targeting cost control and performance, new ScyllaDB X Cloud capabilities include faster data streaming, mixed node sizes within clusters to simplify operations and improved throughput that enables ScyllaDB to process more data in a given period of time.

Improved storage utilization -- when used in conjunction with autoscaling -- is perhaps the standout new feature given that it addresses infrastructure costs, according to Aslett.

"Improved resource utilization should enable enterprises to lower infrastructure costs by avoiding over-provisioning, especially with the autoscaling capability that enables users to define utilization targets against which scaling to be automatically triggered and optimized," he said.

Pratt, meanwhile, highlighted extending tablet-based elasticity to new workload types and improved storage utilization.

"ScyllaDB's message here is operational predictability, scaling and efficiency that are consistent enough to plan and budget around," he said.

From a competitive standpoint, the new features align ScyllaDB with competitors such as Apache Cassandra, DynamoDB and DataStax -- now part of IBM -- Pratt continued.

Next steps

ScyllaDB has always taken a close-to-the-metal approach to designing its platform that prioritizes direct interaction between the database system and its underlying hardware while using the C++ programming language to build its architecture, according to Laor.

With the latest X Cloud update now generally available, he said that ScyllaSB's next initiatives are to improve the consistency of the platform's data plane and to add tiered storage such as "hot storage" that enables low-latency access to frequently used data, "warm storage" with medium performance for less frequently used data and "cold storage" at a low cost for rarely used data.

Something else ScyllaDB could focus on is adding prowess to its vector search and storage capabilities, according to Aslett.

Beyond vector database specialists such as Pinecone and ChromaDB, many data management vendors including Databricks and Snowflake added vector search and storage in 2023 and 2024 to help customers suddenly interested in developing AI applications following OpenAI's late 2022 launch of ChatGPT.

ScyllaDB, however, only added such capabilities as part of its ScyllaDB 2025.4 release on Jan. 8; ScyllaDB 2025.4 is the self-managed version of the vendor's platform while ScyllaDB X Cloud is its fully managed version.

"Data platform providers seeking to differentiate in relation to vector storage and retrieval can still do so by focusing on vector indexing as a means of improving the performance and accuracy of similar search results, as well as implementing hybrid search capabilities," Aslett said.

Pratt, meanwhile, suggested ScyllaDB simplify migration workflows to make its database easier for new customers to adopt.

"ScyllaDB is moving in the right direction by combining operational simplicity with modern capabilities, and the next step is making adoption even easier for more teams," he said.

Eric Avidon is a senior news writer for Informa TechTarget and a journalist with more than 25 years of experience. He covers analytics and data management.