“AWS can design everything from the silicon in its servers to the embedded OSs in edge devices and the complete stack of software in between.”Gartner
It has been over a decade since its launch, and today AWS has grown into one of the most successful cloud infrastructure companies on the planet, garnering more than 30 percent of the market. AWS still remains to be one of Amazon’s strongest units, accounting for 77% of Amazon’s total operating profit for the quarter. Last year, AWS generated $35 billion in revenue for the company. In the latest Gartner announcements, AWS has been named the leader again, for the 10th straight time.
AWS has been consistent with meeting the fast-transitioning demands of the developers. Especially with the outburst of machine learning use cases, customers have wanted to mount their ML workloads on cloud platforms. Let’s take a look at how AWS has fared in terms of providing ML services.
Traditional ML development is complex, expensive and iterative. Finding the right tools and integrating them can be time-consuming and error-prone. SageMaker solves this challenge by providing all of the components used for machine learning in a single toolset so that models get to production faster, with less effort.
For example, using SageMaker Autopilot, one can automatically inspect raw data, apply feature processors and pick the best set of algorithms, track their performance all with just a few clicks.
The more efficient the code and application is, the less costly it is to run. Developers can integrate Amazon CodeGuru into existing workflows, and the built-in code reviews detect and optimise the code to cut down costs. Amazon CodeGuru uses ML to provide intelligent recommendations for improving code quality and identifying an application’s most expensive lines of code.
CodeGuru Profiler provides visualisations and recommendations on how to fix performance issues and the estimated cost of running inefficient code. Whereas, CodeGuru Reviewer uses machine learning to identify the critical problems and hard-to-find bugs during application development.
Identifying the objects and scenes in images that are specific to your business needs the use of Rekognition. While building models to monitor assembly lines or farms, Rekognition makes it easy to add image and video analysis using highly scalable, deep learning technology that requires no machine learning expertise to use.
Amazon’s Comprehend is an NLP service that uses machine learning to find insights and relationships in a text. With no machine learning experience required, users can leverage AutoML capabilities in Comprehend to build text classifiers that suit the needs of an enterprise.
According to AWS, inference in machine learning makes for 90% of the total operational costs. Using Elastic Inference, one can reduce the cost of running deep learning inference by up to 75%. Developers just have to attach the right amount of GPU-powered inference acceleration to any of the services such as EC2 or SageMaker instance type or ECS task, without the need for changing code.
Make searching easy on your website with Amazon Kendra. Powered by machine learning, Amazon Kendra is a highly accurate and easy to use enterprise search service. Kendra delivers powerful natural language search capabilities to websites and applications so that the end-users can find the information easily.
Augmented AI (A2I)
To any organisation, building human review systems is a time consuming and expensive process. By removing the heavy lifting associated with building human review systems, Amazon Augmented AI makes it easy to build the workflows. Amazon’s A2I service also makes it easy to integrate human judgement and AI into any ML application, regardless of whether it’s run on AWS or on another platform.
Chatbots, podcasts or home pods, having a human-like voice delivery is a huge challenge for B2C companies. With Polly, AWS tries to address the same. Customers can use this service to deliver lifelike voices and enhance user experience with real-time output. All one has to do is to send a text to Amazon Polly’s API and return the audio as a stream, which can be played immediately. Amazon has equipped Polly with the ability to generate speech in dozens of languages, making it easy for the global audience.
All these tailor-made services have been powering organisations in various ways across the globe. AWS has even forayed into high-performance sporting events such as Formula One. F1 analysts and AWS have partnered to draw deep insights out of each moment and present it to the viewers in an intuitive way. The applications of AWS have been extended to the creative space as well, such as AWS DeepComposer can get you started with playing keyboard within seconds.
Even though the race has become much tighter in the years due to the acceleration of Microsoft and Google’s offerings, and the evolution of other cloud platforms, AWS has still managed to rule the rooster.
If you loved this story, do join our Telegram Community.
Also, you can write for us and be one of the 500+ experts who have contributed stories at AIM. Share your nominations here.
I have a master's degree in Robotics and I write about machine learning advancements. email:firstname.lastname@example.org