AWS looks to take the drudge work out of data analysis
- 02 December, 2016 10:40
Amazon Web Services is looking to make it easier, and more efficient, for enterprises to analyze their data in the cloud.
"Eighty percent of what we call analytics is not analytics at all but just hard work," said Werner Vogels, chief technology officer at Amazon.com, speaking during a keynote speech this morning at the AWS re:Invent cloud conference in Las Vegas.
Instead of digging down into a company's data to find patterns and insights that will give an enterprise a competitive advantage, too much time is spent on indexing, storage, security, and making sure the right access is set up.
To help AWS cloud customers with that analysis workload, the company today unveiled AWS Glue .
The new service is designed to be a fully managed data catalog and ETL (extract, transform, load) system. It's all designed to make it easy to move data between numerous databases, apps, and systems, while also making it easier and faster to handle data discovery, mapping, and job scheduling tasks.
AWS Glue has not been released and Amazon would not put a release date on it, with its website saying only that it's "coming soon."
According to the vendor, AWS Glue is being integrated with Amazon S3, Amazon RDS, and Amazon Redshift, and can connect to any Java Database Connectivity -compliant data store.
Glue is being built to cull a company's data stores, picking out different formats and then suggesting specific schemas and transformations, so IT workers don't have to do all of that work by hand.
That, says Vogels, will leave people more time to do actual analysis, rather than managing the background workload.
"We are really flipping this around for you," Vogels told the keynote audience. "Eighty percent of the work you do will be analytics -- truly analytics -- and only 20% will be just work."
One analyst said it could be a useful tool.
"Using Glue could help enterprises get insight into their data," according to Patrick Moorhead, an analyst with Moor Insights & Strategy. "It's essentially crawling around, looking for patterns in the data that the enterprise may not know of."