While scaling out the database is the norm for Big Data systems, there are many hidden costs and complexities associated with “node sprawl” that can be remedied by instead deploying smaller clusters of larger database nodes. This paper explores the merits of small clusters and shares test results that debunk the concerns some organizations have about using large nodes.
In this paper we explain the key design decisions that went into building a drop-in replacement for Apache Cassandra with scale-up performance of 1,000,000 IOPS per node, scale-out to hundreds of nodes and 99% latency of less than 1 millisecond. Accomplishing this called for rethinking many of the foundational architectural choices behind other NoSQL databases.
In this paper, we compare Scylla with Amazon DynamoDB. We’ll cover the detailed methodology of our testing before demonstrating that Scylla performs significantly better than Amazon DynamoDB under real-world conditions. Our evaluation also demonstrates that Scylla delivers significant cost savings over Amazon DynamoDB.
With the release of Scylla Open Source 3.0, we’ve introduced a rich set of new features for more efficient querying, reduced storage requirements, lower repair times, and better overall database performance. Already the industry’s most performant NoSQL database, Scylla now includes production-ready features that surpass the capabilities of Apache Cassandra.
Over-eager ingestion can result in a buildup of queues of background writes, possibly to the point of depleting available memory.This paper explains how Scylla ensures that ingestion of data proceeds as quickly as possible, but not quicker.