Following what struck me as a comparatively dry spell of products bulletins in 2021, AWS spent re:Invent 2022 launching a host of new services. AWS Main Evangelist Jeff Barr, with support from some AWS developer advocates, summarized the most impactful bulletins because “there’s simply also considerably wonderful things for the team to deal with,” but then they proceeded to spend much more than 2,700 phrases highlighting their beloved announcements, which seemed to include… anything. Mainly, they handed out participation trophies to each and every AWS service staff. Not significantly practical.
They could have highlighted automated knowledge preparation for Amazon QuickSight Q, supplied how difficult knowledge preparing can be for equipment learning. Or what about Amazon Stability Lake, which mechanically centralizes a company’s safety facts from cloud and on-premises resources into a info lake. Extremely interesting. What about Amazon CodeCatalyst, which RedMonk analyst James Governor rightly characterizes as “a packaging exercise” intended to boost application growth and delivery and guide to greater convenience (“the killer app”). Also, incredibly great.
If we look further than the gazillion new companies and updates to current products and services that AWS announced, an rising concept portends a significantly unique (and much better) AWS. Yes, I’m chatting about integration as an critical merchandise function.
Less assembly expected
AWS utilised to tout its 200+ services. Not any more. In reality, there are probably closer to 400 AWS providers now, but at some issue in the past two many years, AWS recognized that possessing so quite a few solutions challenging customers’ IT decisions fairly than simplifying them.
For those people unfamiliar with how AWS operates, each individual assistance (solution) team runs autonomously. There is some leading-down direction, but as a standard rule, particular person provider groups make what they feel customers most want, even if that sales opportunities to inter-team competition. This is both of those a attribute (autonomous teams can establish a lot quicker) and a bug (autonomous teams really do not essentially coordinate to make it quick to use a number of AWS products and services harmoniously). Clients are normally still left to cobble jointly disparate companies without the need of tight integration in the way Microsoft could possibly deliver, for example.
All this makes the introduction of Amazon Aurora zero-ETL integration with Amazon Redshift this sort of a jaw-dropper.
Let us be apparent: In essence, AWS introduced that two of its products and services now perform well collectively. It’s more than that, of system. Taking away the expense and complexity of ETL is a wonderful way to clear away the will need to build data pipelines. At coronary heart, this is about producing two AWS providers get the job done extremely perfectly alongside one another. For an additional corporation, this may possibly be deemed table stakes, but for AWS, it is comparatively new and extremely welcome.
It’s also a signal of where AWS could be headed: tighter integration between its individual providers so that clients needn’t consider on the undifferentiated weighty lifting of AWS provider integration.
Earning place for third parties
That zero ETL announcement, as powerful as it was, would have been even superior had AWS also highlighted seamless integrations with 3rd-celebration expert services this sort of as Databricks or DataStax. AWS could not like to use the “P” phrase (“platform”), but that doesn’t transform reality. AWS is the world’s largest cloud system, and AWS clients rightly count on to be in a position to combine their most well-liked application with AWS.
This is what can make Amazon DataZone so attention-grabbing.
Amazon DataZone is a “data administration service that aids you catalog, find out, analyze, share, and govern facts throughout the corporation,” writes Swami Sivasubramanian, AWS vice president of info and equipment learning. This would be cool if all it did was pull collectively all the facts stored in repositories from a variety of AWS expert services, which it does with integrations to AWS expert services like Redshift, Athena, QuickSight, and extra. DataZone goes past this by featuring APIs to combine with 3rd-bash info sources from associates or others.
On the one particular hand, it is clear that AWS would have to offer this sort of APIs, mainly because of course, not all (or even most) customer data sits in AWS. In the FAQ accompanying the announcement, AWS even outlined that DataZone can keep track of knowledge in rival cloud companies like Google Cloud and Microsoft Azure—multicloud, anybody? But it is also not obvious. Just after all, the tech marketplace has invested many years viewing Apple, Microsoft, and some others disregard aggressive goods outside their personal walled gardens. By emphasizing the have to have to obtain non-AWS facts resources, DataZone may nicely be a leading indicator of AWS going over and above grudging acceptance of third-get together knowledge sources or expert services to emphatic embrace.
Then there was the announcement that wasn’t an announcement at all. AWS announced Dependable Language Extensions for PostgreSQL on Amazon Aurora and Amazon RDS. PG.TLE is an open supply growth kit for setting up PostgreSQL extensions. It “provides databases directors command in excess of who can put in extensions and a permissions model for operating them, allowing software builders provide new functionality as soon as they figure out an extension meets their requires.”
What was not declared and by no means will be is the simple fact that AWS is arguably the second-premier employer of PostgreSQL contributors, just at the rear of CrunchyData. I’ve suggested before that AWS has increasingly witnessed the need to lead to the open source assignments on which its managed providers (and its prospects) count. AWS staff contributions to PostgreSQL are a robust example of this.
All of this implies that AWS is starting to be less insular each day. The organization has usually considered “customer obsession” as its most important achievement metric, and at times provider groups felt the proper way to achieve that was to establish the most effective achievable support in isolation from the customer’s existing IT investments, which includes other AWS products and services. It also led some teams to limit their involvement in upstream open up resource jobs and try to deliver a self-contained variation of that challenge so as to greater regulate the buyer knowledge.
As these and other re:Invent announcements suggest, AWS increasingly builds community—whether partners, open resource tasks, or even other competitive products—into its expert services. That’s fantastic for clients and good for AWS.
Copyright © 2022 IDG Communications, Inc.