Friday, December 1, 2017

The AWS rocket continues its ascent

The annual AWS re:Invent 2017 conference in Las Vegas provided a big stage for the world’s biggest cloud company and maybe 40,00o of its best friends, customers, and erstwhile ecosystem partners whose business models have not yet been subsumed by the power that is Amazon.

Andy Jassy, who heads Amazon Web Services, is certainly setting high ambitions and at the same time boasting of the substantial lead the company is building over Microsoft Azure, Google Cloud Platform and especially Oracle, which still fancies itself a close competitor despite lacking the hyperscale data centers that the top three competitors are assembling worldwide.

Like many keynotes, Jassy’s three-hour presentation on Wednesday introduced a new set of marquee customers who are in the process of migrating their IT infrastructure into the AWS embrace.  Some big names include Expedia, the National Football League, and Disney, which has long been using AWS for various cloud workloads but which has now designated AWS as its preferred public cloud infrastructure provider.

As for ecosystems partners, everyone wants to jump on this wagon. Amazon Marketplace network security partners highlighted during the keynote included Cisco, Juniper, Palo Alto Networks, Fortinet, Barracuda, Trend Micro, Sophos, Tenable, F5 and Imperva.  AWS also recently announced Symantec as another key partner – one who is going “all in” by moving workloads to the AWS infrastructure. But beware! AWS recently launched its own security monitoring service – just like how it is now offering a deep set of video encoding services under its own label. Will this become a recurring pattern? First, encourage a wide ecosystem but then introduce your own service?

Here’s a few of the recent AWS service launches:

Amazon Neptune – a new graph database service that spurs insights from relationships among highly connected datasets. The core of Amazon Neptune is a purpose-built, high-performance graph database engine optimized for storing billions of relationships and querying the graph with milliseconds of latency.  The service is delivered as a fully managed database and runs within an Amazon Virtual Private Cloud, enabling data in transit and at rest to be encrypted.

AWS Fargate – a new way to deploy containers on AWS. Fargate is like EC2 but instead of a virtual machine you get a container. Fargate uses existing ECS primitives, APIs, and AWS integrations. Fargate provides native integrations with Amazon Virtual Private Cloud, AWS Identity and Access Management (IAM), Amazon CloudWatch and load balancers.

Amazon Time Sync Service – this is a time synchronization service delivered over Network Time Protocol (NTP) which uses a fleet of redundant satellite-connected and atomic clocks in each region to deliver a highly accurate reference clock. AWS will provide this service at no additional charge in all public AWS regions to all instances running in a VPC.

Amazon FreeRTOS -  an IoT microcontroller operating system that simplifies development, security, deployment, and maintenance of microcontroller-based edge devices. Amazon FreeRTOS extends the FreeRTOS kernel, a popular real-time operating system, with libraries that enable local and cloud connectivity, security, and (coming soon) over-the-air updates.

AWS IoT Analytics - a fully managed service that provides analysis of data collected from IoT devices

Amazon Translate - a neural machine translation service that uses machine learning to provide fast language translation of text-based content and enable the development of applications that provide multilingual user experiences. The service is currently in preview and can be used to translate text to and from English and the supported languages.

Amazon Transcribe - an automatic speech recognition (ASR) service that lets developers add speech to text capabilities to their applications.

One other footnote:

  • Equinix announced the extension of direct, private connectivity to the AWS Direct Connect service to four additional data centers in Helsinki, Madrid, Manchester and Toronto, bringing 

In his keynote address,  Amazon CTO Werner Vogels talked about how AWS is democratizing access to next gen technologies especially machine learning and AI. Amazon is certainly bringing down the cost of centralized processing and storage by leveraging the power of scale and extreme efficiencies.
Democratizing is a tricky adverb as it implies a feeling of egalitarianism, or at least that the outcome will be for the public good. Too soon to say that, but cleary Vogels has an ambition to make cloud services more human-centric.

AWS is quickly moving up the stack to support image and speech processing as the cloud-enabled GUI for next gen applications. There is an urgency to "liberate" workloads, meaning to decouple workloads from on-premise servers so that they will flow seamlessly onto the AWS cloud. In terms of market dominance, AWS has released a steady stream of Fortune 500 enterprises declaring them to be "the preferred cloud infrastructure provider." The company is now claiming a 44.1 percent share of the global cloud market, compared to its next nearest competitor -Microsoft -at 7.3 percent.

A few notes:

New GPU-powered EC2 instances -- the new P3 instances offer up to eight NVIDIA Tesla V100 GPUs and are designed to handle compute-intensive machine learning, deep learning, computational fluid dynamics, computational finance, seismic analysis, molecular modeling, and genomics workloads.Each of the NVIDIA GPUs is packed with 5,120 CUDA cores and another 640 Tensor cores and can deliver up to 125 TFLOPS of mixed-precision floating point, 15.7 TFLOPS of single-precision floating point, and 7.8 TFLOPS of double-precision floating point.

Amazon Rekognition Image -- a cloud service that uses deep learning to provide scalable image recognition and analysis. It leverages deep learning techniques to build and integrate object and scene detection, real-time facial recognition, celebrity recognition, image moderation, as well as, text recognition into applications and systems. Amazon developed the service by using its archives of its own Prime Photos to analyze billions of images each day. Examples shown included large scale facial recognition to identify faces in a crowd by comparing to social media sources.

Caltech - AWS is forming a research partnership with Caltech focused on AI, data science, and machine learning. Researchers from the Computing and Mathematical Sciences (CMS) and Electrical Engineering (EE) departments at Caltech will use the AWS Cloud (including Nvidia GPU instances) to train deep neural networks using open source projects like Apache MXNet. Of course, Amazon will bring financial support.

AWS Cloud9 - a web-based Integrated Development Environment (IDE) for writing, running, and debugging code. Cloud9 was developed by a start-up of the same time, which Amazon acquired last year. It supports multiple programming languages, including Javascript, Python, and PHP. AWS expects Cloud9 to provide a seamless experience for working with serverless applications.

AT&T and AWS are working on a cellular LTE-M Button that utilizes the newly announced AWS IoT 1-Click service. The idea is for enterprises and developers to integrate their operations/workflows with AWS IoT, AWS Lambda, Amazon DynamoDB, Amazon Simple Notification Service (Amazon SNS) and many other AWS solutions. For example, pressing the button could trigger the order of office supplies or the submission of a customer service request.  AT&T will offer the LTE-M button at $29 for the first 5,000 units. The button will be released in early 2018. Prices will be higher after that. This will help in scaling IoT. It also appears to tweak the business model a bit. Apparently, no data charges? No monthly charges? The click of a button obviously does not put a strain on the LTE network. If the battery is fixed, perhaps the expectation is that the button will function for 48 months and then deactivate itself when the battery dies.

See also