Serverless vs self-management

Serverless vs self-management – ​​Gigaom


NoSQL databases and relational databases store different data, and are divided into several different “types” based on their data models. As almost all companies have a huge demand for data, NoSQL databases have shown strong growth in the past few years, and it is estimated that by 2025, the global NoSQL market will grow by as much as 30%.

But the undisclosed cost of ownership of NoSQL databases can place an unsustainable burden on many organizations. To help IT decision makers understand the problem, GigaOm analysts William McKnight and Jake Dolezal released a new report, Cassandra total cost of ownership research, It examines the full cost and true value of the popular Cassandra NoSQL database self-managed on Google Cloud and the fully managed DataStax Astra serverless Cassandra product.

The total cost of ownership (TCO) for the last three years found that the cost of serverless Cassandra was $740,087, while the cost of a self-managed OSS Cassandra deployment was $3,174,938. Researchers say this is an amazing difference.

According to McKnight, the serverless option “provides what is necessary as needed, thereby eliminating the complex and costly over-construction that would otherwise be caused.”

He admitted that he was surprised how much cost savings a serverless Cassandra deployment could save, and found that compared with self-management, the cost was reduced by 76%, the personnel cost was reduced by 95%, and the complexity was reduced by 3 times. The architecture cost has been reduced by 44% instead of the self-managed Cassandra.

McKnight said: “In the past, I conducted more rigorous research on customer situations, compared open source and commercial options, and came to similar conclusions, although the gap is not as obvious as it is now.” OSS Cassandra’s operating costs have been reduced by 44% and surprised.”

Analysts pointed out that although cost is important, it is only one factor in choosing the Cassandra platform. There are many other aspects to consider, including performance, management, functions and features, workload management, user interface, scalability, vendors, reliability, and many other conditions. However, studies have shown that serverless Cassandra can significantly reduce deployment time and costs compared to self-managed Cassandra workloads.

Related Articles

Back to top button