enterprise| enterpriseauthorTom KrazitNoneAre you keeping up with the latest cloud developments? Get Tom Krazit and Joe Williams' newsletter every Monday and Thursday.d3d5b92349
×

Get access to Protocol

I’ve already subscribed

Will be used in accordance with our Privacy Policy

The New Enterprise

How Zoom pulled off the scaling event of a lifetime

Self-managed data centers with excess capacity, an application built to adapt to bandwidth and key cloud alliances all helped Zoom weather the work-from-home storm.

A very large Zoom call

Zoom was able to handle that surge in demand this year with minimal service disruption, thanks to a global hybrid cloud network of self-managed data centers and cloud resources.

Image: Zoom/Protocol

All modern enterprise tech companies plan for scale. Few have been tested to the degree that Zoom was at the outset of the worst pandemic in modern history.

In December 2019, 10 million people a day participated in a meeting on Zoom. By March 2020, that number had grown to 200 million as governments imposed lockdowns around the world in hopes of containing the spread of COVID-19. And during the 90 days between Feb. 1st and April 30, Zoom experienced a 224% increase in the number of customers with 10 or more employees using its product compared to the previous 90-day period.

By and large, Zoom was able to handle that surge in demand with minimal service disruption, thanks to a global hybrid cloud network of self-managed data centers and the cloud resources of AWS and Oracle.

"There's a lot of things in terms of the way that we design the platform that enabled us to really scale in this situation," said Brendan Ittelson, chief technology officer at Zoom, in an interview with Protocol. "If we need additional data center capacity, we have the ability to quickly deploy it, and ramp up in either our data centers or with our public cloud providers so that we can scale that way."

Zoom's videoconferencing application was designed to make use of multi-bitrate encoding, which allows the application to provide different levels of video quality depending on local network conditions and demand, Ittelson said. That meant Zoom could still provide service to homes around the world that didn't necessarily have the same types of high-bandwidth video connections commonly found in commercial office districts without falling over, even if the quality wasn't superb.

For the most part, the company uses its network of 19 co-located data centers around the world to service meeting traffic, Ittelson said. Going into 2020, those data centers were running at about about 50% of peak capacity, he said, which gave Zoom the ability to absorb the early impact of increased demand in China and Europe as the pandemic spread.

Zoom's goal all along was to provide its services as close to the end user as possible through these data centers. Distance is an extremely important factor for video applications, given that stuttering video feeds and audio gaps caused by latency can ruin a big meeting in a hurry.

But Zoom is also a longtime customer of AWS, and uses the cloud provider's network for other parts of the Zoom application. During a video discussion in April, Zoom CEO Eric Yuan said that before the pandemic, Zoom was using AWS for pre-meeting and post-meeting activities, but started using AWS servers for meeting traffic as the impact of the pandemic became clear.

And Zoom customers aren't using the service for meetings alone.

"Zoom is a pretty wide product offering," Ittelson said. "We have the meetings, webinars, our Zoom Phone product, our Zoom Chat product … naturally, [in] the infrastructure for each of those, we use a range of components. There is everything from virtualization to bare metal in these different environments."

Zoom added Oracle as a cloud supplier in April as demand for its service skyrocketed and Oracle dangled price incentives, but Ittelson declined to talk about how Zoom assigns workloads between the two cloud providers, who don't like each other very much.

In November, Oracle co-founder and CTO Larry Ellison claimed that "the majority of Zoom meetings — these tens of millions of Zoom meetings that happen every day — happen in the Oracle cloud," which doesn't square with Ittelson's description of the role Zoom's own data centers play when it comes to serving meeting traffic. Both AWS and Zoom CEO Eric Yuan have also disputed Ellison's claims about Oracle's role inside Zoom's infrastructure strategy.

Buoyed by a surge in revenue during the first half of 2020, Zoom is planning on reinvesting that windfall in its own data center capacity over the coming months, Yuan said on the company's second-quarter earnings call. Zoom has added data centers in Osaka, Japan, and Singapore since the initial surge in February and early March, and announced plans back in September to invest in the capital expenditures required to increase that footprint.

As Zoom nears the end of a year that few of us will ever forget, it's being cautious about what it expects for 2021, during which a COVID-19 vaccine is expected to become available and life in North America and Europe will hopefully start resembling something normal.

While it's unlikely that Zoom will experience the type of all-hands-on-deck scaling event it saw in the first part of 2020, it's also unlikely that demand for videoconferencing services will fall off a cliff. Microsoft co-founder Bill Gates told attendees of the virtual New York Times DealBook conference in mid-November that he expects business travel will fall by half over the foreseeable future. That will make for lots of additional virtual meetings to come — and probably plenty more opportunity for Zoom to grow.

"I think there's a new range of things that will come out, and we're not going to go back to the way it was," Ittelson said. "We're excited to see where folks continue to shift."

Correction: A previous version of this story used the wrong metric for Zoom's use. It measures use via daily meeting participants, not users. Updated Dec. 10, 2020.

The New Enterprise

Software ate the world. Now it’s eating software companies.

Enterprise software juggernauts are trying to emulate (or just acquire) the smaller, specialized companies. It's not as easy as it looks.

Nom nom nom.

Image: Bloomberg/Getty Images/Protocol

Do you want excellence, or convenience?

Cloud computing allowed a generation of enterprise software companies to flourish, providing them the freedom to tinker with laser-focused ideas that could streamline stodgy business workflows within a browser or an app. That has led to a proliferation of different enterprise software vendors offering high-performance but niche products: According to data from Okta, which makes identity management software, its average corporate customer is using 88 separate apps across their business.

Keep Reading Show less
Tom Krazit

Tom Krazit ( @tomkrazit) is a senior reporter at Protocol, covering cloud computing and enterprise technology out of the Pacific Northwest. He has written and edited stories about the technology industry for almost two decades for publications such as IDG, CNET, paidContent, and GeekWire. He served as executive editor of Gigaom and Structure, and most recently produced a leading cloud computing newsletter called Mostly Cloudy.

The New Enterprise

Nine companies that could define the future of enterprise software

From streamlining software development to changing the way companies communicate, here are the software trends headed to your offices in the coming years.

A bunch of enterprise software companies want companies to do things a little differently.

Image: Protocol

Software may have eaten the world, but the world has changed — and a new wave of enterprise software has an appetite.

Heading into 2021, the enterprise software industry has a wealth of opportunity before it, not least as a result of COVID-19. As companies raced to make better use of cloud computing, their needs shifted: They now need to build, deploy and monitor software in wholly new ways. At the same time, companies are also using digital systems more than ever — whether that's for customer engagement, internal communication, staff training or something else entirely. And on top of all that is the natural march of technological progress as technologies like AI and VR mature and finally become usable in the workplace (or home office).

Keep Reading Show less
Hirsh Chitkara
Hirsh Chitkara (@ChitkaraHirsh) is a researcher at Protocol, based out of New York City. Before joining Protocol, he worked for Business Insider Intelligence, where he wrote about Big Tech, telecoms, workplace privacy, smart cities, and geopolitics. He also worked on the Strategy & Analytics team at the Cleveland Indians.
The New Enterprise

AWS quietly enters the multicloud era

The cloud leader confirmed that two new services introduced Tuesday at re:Invent can be used to manage applications on Microsoft and Google's cloud services.

AWS CEO Andy Jassy gave a keynote at AWS' re:Invent.

Photo: AWS re:Invent

It was easy to miss a line on two slides presented during AWS CEO Andy Jassy's Tuesday morning keynote at re:Invent 2020. But it opened a new chapter in the history of the cloud computing pioneer.

ECS Anywhere and EKS Anywhere — two new versions of AWS' managed containers and managed Kubernetes services, both designed for customer data centers — can be used to manage applications running on Microsoft Azure and Google Cloud. Each service "works on any infrastructure," according to the slide, and AWS confirmed to Protocol that they would allow customers to use its software to manage workloads running on other cloud providers.

Keep Reading Show less
Tom Krazit

Tom Krazit ( @tomkrazit) is a senior reporter at Protocol, covering cloud computing and enterprise technology out of the Pacific Northwest. He has written and edited stories about the technology industry for almost two decades for publications such as IDG, CNET, paidContent, and GeekWire. He served as executive editor of Gigaom and Structure, and most recently produced a leading cloud computing newsletter called Mostly Cloudy.

The New Enterprise

Inside S&P Global Ratings’ aggressive computing overhaul

Mark Wang, the company's head of cloud engineering, had a three-year plan to reboot its approach to computing. It was ambitious to say the least.

"We've built out all the cloud expertise in-house, that's one thing I'm very proud of," says Mark Wang, the company's head of cloud engineering.

Image: Creative-Touch/Protocol

"We're a 160-year-old institution," says Mark Wang, head of cloud engineering at S&P Global Ratings. "Now, we're moving at the pace of a fintech."

Unlike many fellow financial services companies, which have been slow to adopt cloud computing, Wang's team at the global ratings agency has been nothing less than extremely aggressive in its rollout of new technology. Last year, it moved more than 160 of its internal applications to the cloud, bucking the trend of compromising on the hybrid cloud. This year, it's embarking on an ambitious plan to re-architect those applications around serverless computing principles and Kubernetes, using the Knative open-source project.

Keep Reading Show less
Tom Krazit

Tom Krazit ( @tomkrazit) is a senior reporter at Protocol, covering cloud computing and enterprise technology out of the Pacific Northwest. He has written and edited stories about the technology industry for almost two decades for publications such as IDG, CNET, paidContent, and GeekWire. He served as executive editor of Gigaom and Structure, and most recently produced a leading cloud computing newsletter called Mostly Cloudy.

The New Enterprise

Data centers aren’t dead. But they’ll never look the same again.

Major cloud providers have accepted a future where hybrid cloud strategies take center stage, which means on-premises infrastructure will continue to evolve. Here's how.

All-in cloud strategies are likely to be the exception, rather than the rule, when it comes to the next wave of modernization.

Photo: Erik Isakson/Getty Images

A funny thing happened along the way to the cloud-only future we were promised a decade ago: Turns out, for many applications, the old-fashioned way of doing business on the internet works just fine.

Cloud vendors and customers have reevaluated their infrastructure strategies around the hybrid cloud in recent years, and that means self-managed data centers are going to be with us for a long time to come. Even AWS, the pioneer of the cloud market, now offers customers a rack of physical servers designed around AWS services that they can use in their data centers, just a few years after scoffing at the notion that anyone would want to manage their own equipment.

Keep Reading Show less
Tom Krazit

Tom Krazit ( @tomkrazit) is a senior reporter at Protocol, covering cloud computing and enterprise technology out of the Pacific Northwest. He has written and edited stories about the technology industry for almost two decades for publications such as IDG, CNET, paidContent, and GeekWire. He served as executive editor of Gigaom and Structure, and most recently produced a leading cloud computing newsletter called Mostly Cloudy.

Latest Stories