Building out more regions to support the cloud can further cut down network latency and cost to reach end users, even in space, in near real-time.

Joao-Pierre S. Ruth, Senior Editor

December 7, 2021

4 Min Read
AWS logo on a sign at re:Invent
AWS re:Invent file photoJoao-Pierre S. Ruth

In years prior, AWS CTO Werner Vogels somewhat jokingly said he would “fix the speed of light” as the new world of cloud removed constraints in compute. During his keynote at the 2021 re:Invent conference, he said that task remained on his to-do list, but there were more immediate possibilities within reach.

Vogels revisited some of his comments about innovation from previous re:Invent conferences to frame how past constraints stymied the capacity to develop and grow. “Before cloud, you either had to get massive investments, you had to buy hardware, you had to hire IT people -- things that had nothing to do with actually building a product.”

What cloud did, he said, was take everything that was constrained—all the hardware pieces—and make them all programmable. “Suddenly, getting access to capacity was just a click of a button.”

Data centers and other physical resources, for example, became virtually programmable. The early days of Amazon Elastic Compute Cloud (EC2) saw a simple interface, Vogels said, and since that time, EC2 has become ambitious in developing instances -- virtual servers in the cloud -- to accommodate user demand.

“Every day almost, you asked us for a different instance family,” he said. “You wanted storage-optimized ones. You wanted compute-optimized. You wanted to have large memory instances where you could run your SAP HANA in.”

The first instances were based on a limited hypervisor, he said, predicating a need for massive investment and innovation in data centers to ensure AWS could get next-generation compute platforms into customers' hands. At this year’s conference, Vogels announced new Mac EC2 M1 instances, which would offer greater efficiency. “Apple claims for the M1 Mac instances that there is a 60% cost benefit improvement over the platforms they used before.”

Evolution of Cloud

Though the evolution of the cloud has removed some constraints on innovation and development, he said there is more that can be done from digital and policy perspectives. “There are a number of laws that continuously are constraints we have to deal with,” Vogels said. “Whether there is latency in bandwidth ... or law of the land.”

The latter is an easier matter to address, he said, by working with regulatory agencies to educate them about cloud, its capabilities, and the heightened level of security possible for data in the cloud compared with private data centers.

The expanding possibilities of cloud can reach beyond terrestrial innovation. Payam Banazadeh, CEO and founder of Capella Space, a developer of observation satellites, joined Vogels to speak on the opportunities cloud opens up for working with data transmitted from space.

“Sensors from space have a truly unique vantage point,” Banazadeh said. “They are needed for us to have a connected world.” He said organizations need constant global access to information in order to make decisions on global scale. “For this, space is the missing link,” Banazadeh said. Satellites from Capella Space can capture images and other surface changes on the planet down to millimeter accuracy, he said. “It gives us reliability and visibility into our planet in all conditions. That’s a fundamental requirement to getting to real-time monitoring.”

Banazadeh said he expects Capella’s satellites to accumulate more than 500 petabytes of data in the coming years that must be transmitted to terrestrial computers. “In order to lay the groundwork for integrating our sensors with other sensors, in real-time with low latency and low reactivity, we had to build our business on a completely different foundation than has traditionally been tried before,” he said. Capella needed a distributed network, full automation, real-time processing of data, and immediate scalability. The satellite developer chose AWS in 2020 for its ground station service.

Vogels gave a rundown of AWS’ overall geographic cloud expansion and how that allowed companies to choose which regions they would use based on latency and cost to reach their respective customers. That growing cloud footprint also helped Amazon broaden its services.

“There’s applications we could not have been building if we didn’t have these regions closer to you,” Vogels said. “In 2006, or 2009, or even 2013, Alexa would not be possible.” Low latency is crucial to Amazon’s voice-controlled virtual assistant, making regional availability of the cloud all the more important. “If you don’t get a response back from Alexa within a second, it doesn’t feel like a natural conversation,” he said.

Advances in hardware, GPUs, and machine learning-enabled capabilities of Alexa, Vogels said, but it was still necessary to drive down latency. AWS maintains 25 regions across seven continents currently, he said. “There are nine more regions planned that we will bring on line in the coming two years.”

Related Content:

Nasdaq CEO at AWS re:Invent Talks Cloud’s Impact on FinTech

Cloud Computing Like a Day in a Chocolate Factory for IT Managers

Capella Space Goes with AWS to Handle Satellite Downlinks

About the Author(s)

Joao-Pierre S. Ruth

Senior Editor

Joao-Pierre S. Ruth covers tech policy, including ethics, privacy, legislation, and risk; fintech; code strategy; and cloud & edge computing for InformationWeek. He has been a journalist for more than 25 years, reporting on business and technology first in New Jersey, then covering the New York tech startup community, and later as a freelancer for such outlets as TheStreet, Investopedia, and Street Fight. Follow him on Twitter: @jpruth.


Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights