AWS Lets New Features Fly At Re:Invent

Wednesday's opening keynote at AWS re:Invent -- as has now come to be expected at Amazon's cloud conference -- let loose a host of new services that will beef up the world's largest and most feature-rich cloud.

Andy Jassy, senior vice president of Amazon Web Services, revealed to the 19,000 re:Invent attendees gathered in Las Vegas the diverse set of new features intended to enrich the platform's database capabilities, ease migrations and introduce advanced data visualization functionality.

As almost always the case with Amazon's cloud, the new services made available to partners and customers are derived from technologies that have been deployed and battle-tested internally within the e-commerce giant.

[Related: 7 Things Coming For AWS Partners]

Sponsored post

Jassy ran through the list, starting with QuickSight, a business intelligence service that allows companies to create visualizations and derive analytic insights without doing any complex data modeling.

"As soon as we recognize an AWS customer, we take all their data from an AWS store and move it to a query engine so they can get their first visualization in 60 seconds," he said. BI vendors partnered with Amazon will be able to use their own tools to increase query time.

QuickSight can also automatically select the best visualization by inspecting data types and relationships, he said.

Matt Wood, AWS' general manager of data science, described the new Super-fast, Parallel, In-memory Calculation Engine (SPICE) that will power QuickSight's calculations.

SPICE, partly enabled through advanced hardware innovations, allows partners to plug in through an API and aggregate data for feeding business intelligence tools from across Amazon data stores.

Jassy then told the audience that Amazon wants to make it easier to process data being transmitted by millions of interconnected devices.

To that end, a couple of years ago, AWS released Kinesis Streams as a platform for processing real-time streaming data.

But the most common use case customers' desire is just to send all data from those devices directly into an AWS store, and even that simple functionality requires writing custom applications.

A new service, Amazon Kinesis Firehose, solves that problem by allowing users to load streaming data into an AWS data store with a single API call.

"This is a very different way to get streaming data into AWS, into the cloud," Jassy said of Kinesis Firehose.

A far different way to send data into the cloud is to put it on a truck.

While that may not sound like a high-tech solution, Amazon Snowball, a new service for shipping data via FedEx, solves a major problem faced by solution providers doing large migrations.

Snowball builds off Import/Export, offering an ultra-tough shippable storage appliance -- a 50 Tb hard drive encased in a tamper-proof enclosure with data encrypted end-to-end and a Kindle on the exterior functioning as an electronic shipping label.

One important technology that has been slow to be embraced in the cloud is the database, Jassy said. At the same time, "it's rare that I meet an enterprise that doesn't want to flee their existing database provider."

Amazon Aurora, revealed last year at re:Invent, addressed that concern, and has overtaken Redshift as the fastest growing service in AWS, used by customers like Expedia, GE, PG&E and NBC Universal, according to the company.

Taking another step in offering customers "freedom from bad database relationships" -- a thinly veiled barb at Oracle -- Amazon launched its sixth database: a fork of open-source MySQL called MariaDB.

Customers want to move as fast as they can to open-source database technologies, Jassy said, but it's difficult to get the same performance. MariaDB expands the possibilities in making that transition, he said.

And since moving on-prem databases to AWS is often an arduous affair, Amazon is also releasing in preview AWS Database Migration Service.

The service makes it much easier to maintain continuity and avoid downtime while migrating a production database, he said.

"It allows you to continuously replicate the data from the source to your new target in AWS," Jassy said, and can track progress with real-time monitoring.

AWS is also launching AWS Schema, a free conversion tool for migrating between database engines.

Jeff Aden, executive vice president of marketing and strategic development at 2nd Watch, said the database-related announcements -- and especially Schema -- were the most interesting and relevant to his company, which may be Amazon's largest channel partner.

"One of the hardest things in migrating a database is not actually migrating the data," he said. "When you go from Oracle to AWS, you have to tweak that stuff, and that can be very time-consuming."

Jassy went on to announce AWS Config Rules, a compliance engine that makes it so "you don't have to make a decision about moving fast or securing the cloud."

Config Rules allows users to quickly implement compliance rules for resource configurations, as well as define actions if those rules aren't being met.

He also announced Amazon Inspector, a service that performs automated assessments that identify security or compliance concerns when deploying applications on AWS.

Inspector, released in preview, looks at the potential impact of a common set of security vulnerabilities on networks, VMs, operating systems and attached storage, and provides detailed reporting about where there might be weaknesses or vulnerabilities.

2nd Watch's Aden said Inspector is a great addition to the cloud platform that should simplify life for partners who must address security as a top customer concern by "automating what's not a service or part of a solution" they offer customers.

Omar Abbosh, chief strategy officer for IT consulting giant Accenture, also joined Jassy on stage at one point during Wednesday's keynote to introduce a new Accenture AWS Business Group.

Accenture hopes the new division, which involves a heightened level of cooperation with Amazon, will help large enterprises complete data and application migrations to the cloud, as well as launch new processes, especially around complex big-data analytics and security goals.

The goal is to compile "a set of capabilities and offerings that were born in the cloud," Abbosh said. "Today is the first step."