AWS Redshift Feels the Heat
Amazon Web Service’s Redshift data warehouse service is taking a beating this week, with database analytics competitors Oracle Corp. and Cloudera claiming their platforms run much faster for less money. To back up its claims, Cloudera released benchmark results said to show advanced capabilities for cloud-native workloads running on its analytics platform as well as improved price performance compared to Amazon Redshift.
Cloudera’s analytics database platform also runs on Apache Impala, an open source incubator project designed to develop a distributed SQL query engine for Hadoop. In benchmark results released on Thursday (Sept. 22), the Palo Alto, Calif., company reported that Impala was 22 percent cheaper than Redshift when querying data stored in AWS (NASDAQ: AMZN) Simple Cloud Storage Service (S3) using cloud native tools.
Impala cost 60 percent less and ran 40 percent faster than Redshift over local EBS storage. It also registered major cost savings and faster time-to-result, Cloudera asserted.
The added horsepower attributed to Impala comes as more business intelligence and analytic workloads move to the cloud. Cloudera argues that the benchmark testing shows that these workloads can tap into the flexibility and cost savings of public cloud services without sacrificing the performance of on-premise analytical databases.
“This comparison [with Amazon Redshift] is clear evidence that Impala is unmatched for these BI and analytic workloads in the cloud,” claimed Charles Zedlewski, Cloudera’s vice president for products.
Impala is touted as decoupling data and computing to provide comparable performance for SQL analytics whether running cloud-natively over data in S3 or across other on-premise and cloud storage options.
The open source tools works natively with data stored on a range of storage engines, including Amazon S3 object store. That, Cloudera said, eliminates the need to move or load data specifically into Impala clusters. “Especially for cloud deployments, this translates to cost-savings and efficiencies as transient clusters can be spun up as needed for BI and reporting workloads and, with cost-effective storage from S3, more data is quickly and readily available for analysis,” the company added in releasing in the benchmark results.
Earlier this week, cofounder and CTO Larry Ellison claimed its Oracle Cloud was more than 100 times faster for database analytics than Amazon Redshift. “Why is Amazon Redshift so slow? Because it’s 20 years behind Oracle,” (NYSE: ORCL) Ellison boasted during a company event this week in San Francisco.
The broadside was part of a larger Oracle campaign aimed at challenging public cloud leader AWS with an expanded analytics package that includes applications, infrastructure and database analytics. Skeptics called Ellison’s challenge to the the public cloud leader “ridiculous.” Added Justin Moore, CEO of data security specialist Axcient: “At the end of the day Oracle will likely be a niche player for certain database and application workloads” running in the cloud.