Vertica outperforms competing cloud analytical platforms in third-party benchmark study
Posted August 7, 2020 by Mike Perrow, Senior Product Marketing Writer/Editor
Choosing a cloud-based big data analytics platform for your organization’s analytical needs is seldom an easy task, and it can’t be the first step on your big data journey. Earlier steps should include research, conversations with your peers, establishing measurable goals, and careful comparisons of big analytics data platforms to learn which one is best suited to your needs.
Along the way, you’ll encounter a lot of hype about this or that cloud-based big data platform. So it can be really helpful when you find information based solely on facts. A new report by McKnight Consulting Group (MCG) is just that kind of resource. MCG used industry-standard benchmarks to test three well-known, cloud-optimized analytical platforms – Vertica in Eon Mode, Amazon Redshift, and a cloud data platform whose vendor didn’t want to be named.
We’re delighted that Vertica came out on top in terms of faster query performance, greater scalability in terms of concurrent users, and less overall cost. We sponsored this benchmark study because we were confident that the other two well-known vendors were slower, offered fewer capabilities and weaker concurrency (i.e., for multiple simultaneous users), and couldn’t handle as many queries per hour compared to Vertica, particularly as data volumes grow from 10 to 50 to 250 TB. It always helps to validate our claim as the unified analytics warehouse with the greatest value for the highest performance.
Take a look at how Vertica stacked up against the competition.
A few details about the McKnight tests
Based on the University of California Berkeley’s AMPLab Big Data Benchmark, specific tests included 1) longest running thread, 2) QPH (Queries Per Hour), and 3) price-performance across 10, 50, and 250 TB of data and concurrency of 10, 30, and 60 users.
UC Berkeley AMPLab provided the pre-existing Big Data Benchmark. MCG accessed the publically available AMPLab BDB data sets via an S3 bucket at s3n://big-data-benchmark/pavlo/. More details about the test setup, query sets, cluster environment, and methodology are in the report.
Highlights of the test results
You have to know before you can crow, and, forgive us, knowing that Vertica came out ahead of the competitors in every instance is worth crowing about. Here’s a snapshot.
- Vertica: Best for scale and concurrency
In all three sizes of data sets, up to 250 TB, and in all three concurrency profiles, Vertica in Eon Mode consistently had the shortest elapsed time for the longest running thread. The following graph compares all three platforms against the 250 TB data set with 30 concurrent users:
- Vertica: The most queries per hour, at every level of scale and concurrency
The impressive results didn’t stop there. Vertica in Eon Mode also completed far more queries per hour (QPH) in every test configuration, up to 5 times more queries than the competitors.
- Vertica: Tops in price-performance costs
Given Vertica’s winning performance and concurrency results, it’s satisfying to know that Vertica in Eon Mode does its work at reduced operational costs compared to the other platforms. Vertica costs 45%-73% less than Amazon Redshift, and 84%-92% less than the unnamed data cloud platform.
Download the McKnight benchmark report today
Using the cloud for big data analytics is a great way to run your variable analytical workloads to help your business stay competitive. But choosing the right platform – one that offers both high-performance, scalability, and affordability – requires validated information based on the numbers, not the hype.
Get your copy of this report and see the comparisons for yourself. Vertica delivers the analytical performance and cost savings that are keeping today’s most data-driven organizations pointed in the right direction.
About the Author
Senior Product Marketing Writer/Editor
Mike Perrow has 25 years as a writer and editor in the software industry, having worked at Powersoft, Sybase, Rational Software, IBM, Hewlett Packard Enterprise, and most recently Micro Focus. He has an extensive background in developing and implementing web and print media, and was the founding editor of The Rational Edge ezine, which published from 2000-2008. He is the co-author with Kurt Bittner and Walker Royce of The Economics of Iterative Software Development, Addison-Wesley, 2008.
Sponsored by Vertica