The Big Story

Complexity in, complexity out

Modern infrastructure computing is "a Scrooge McDuck-level embarrassment of riches," as Redmonk analyst Stephen O'Grady put it in a blog post this week. Never before have so many options been available to anyone trying to build a business on the internet, allowing a wide variety of software to flourish while staying reliable.

But the downside of wide-open possibilities is complexity, and both O'Grady and Austrian computer science professor Daniel Gruss believe that, really, we're asking people to do too much. It's probably time to help developers use and understand enterprise technology the way consumers use and understand consumer technology, which is to say: keep it simple.

• For the last several years, cloud providers have been obsessed with building out a portfolio of services in an attempt to accommodate everything a customer might want.
• This resulted in the aforementioned embarrassment of riches, but also an experience in which the customer pretty much has to figure out how to make everything work together.
• In its last quarterly report, Gartner dinged AWS — the pioneer and leader of basically everything in this market — for creating a library of services that don't always play well together.
• This is leading to what O'Grady believes is a burgeoning opportunity for a cloud provider or enterprise software company to own the "developer experience" by building application development tools that are designed to work together as seamlessly as an iPhone.

Complexity isn't just a productivity issue; it can have enormous and far-reaching security effects.

• The more tools, components and systems people piece together to create some of today's digital wonders, the more exposed they leave those systems to flaws that could bring the whole thing to a halt. What's more, some of these systems are so complex you might not even know something is not secure.
• Gruss, one of the researchers who identified the industry-shaking Meltdown and Spectre chip design flaws, told attendees at Black Hat Asia that computer science needs to think about itself in a fundamentally different way to ultimately solve these challenges.
• Computer science should be studied the same way that natural systems are studied to identify patterns and overlooked developments, he said, now that we've added so many moving parts to our digital works.
• "Our systems are getting more and more complex so we have to invest more and more time into studying them like nature," he said.

While it might take a few years to fundamentally change the way computer science is studied and taught, the market is starting to figure it out.

• Companies built to serve customers managing their own data centers are retooling around software and tools designed to make it easier to get applications up and running on the services that make the most sense for that application, abstracting the hard parts.
• Cloud providers are also doing this, with products like Azure Arc, AWS Outpost and Google Anthos that aim to simplify the experience of modern distributed computing without forcing a team of folks to really understand Kubernetes.
• This could easily be the next battleground for cloud business, especially as a generation of companies unfamiliar with the differences in how the cloud works start to move business out of their data centers.

Most people understand the pleasure of buying and using a well-designed product, whether it's a computer or a toaster (an object that, of course, increasingly comes with a computer). For a long time this just wasn't a priority for enterprise computing: The focus was on proving the value proposition of the cloud, where "insert credit card/get computing resources" seemed simple enough to understand.

That approach worked very well when the cloud was smaller, but it won't work forever, especially as the complexity of the modern digital world increases. The people and companies who figure out how to make customers enjoy using their tools, rather than merely tolerate them, will be in an excellent position to win business from people who like to hear a simple promise: "It just works."

A MESSAGE FROM DIGITAL REALTY

Data Gravity is a Key Megatrend Impacting Enterprise Growth

Digital Realty's new Data Gravity Index DGx™ Report spans industries and geographies. The report helps enterprises understand the dynamics of Data Gravity and how to effectively turn it into an opportunity. Read it here.

This Week On Protocol

APIs FTW? Later on Wednesday the Supreme Court will hear arguments in the decade-long dispute between Google and Oracle over whether or not APIs should be subject to copyright. A victory for Oracle could fundamentally change the way software is built and would place a great deal of uncertainty around nearly two decades of software.

Google's gambit: If you don't like the name of a Google product, just wait five minutes. The company has once again decided to give a well-known product a new name, renaming G Suite to Google Workspace as it builds its own vision of the future of digital work in the post-pandemic era.

Virtual realty: VMware has fared better than several other enterprise tech companies of its generation when it comes to staying relevant in the marketplace, and CEO Pat Gelsinger is one reason why. Coming off last week's VMworld I had a long, wide-ranging conversation with Gelsinger about AI in the data center, Kubernetes hype, and how customers are using VMware's virtualization software in the cloud era.

Five Questions For...

Adam Blitzer, EVP & GM, Digital at Salesforce

What was your first tech job?

I worked for a small advertising agency in Tokyo called Cove-Ito (sadly, it no longer exists). While there, I worked in account management but got a small taste of digital marketing for brands like BMW, Quiksilver and Mini Cooper.

What's the best piece of advice you could give to someone starting their first tech job?

Don't over plan your career. Things can change quickly, and in unexpected ways. Some of the best people I hired at Pardot (acquired by Salesforce) were not from traditional tech backgrounds, but they were curious and hungry; that served them really well. If you lock into a career track, you can miss opportunities just out of your line of sight. Many of the jobs that are in demand right now didn't even exist five years ago.

What was the first computer that made you realize the power of computing and connectivity?

It probably wasn't a computer: It was a modem. It was the first time I was connected to a bulletin board service and connected to other people. It was a very, very, very small version of the internet in some ways. But the first computer I had was IBM XT with a black and orange display, so every day was Halloween.

What was the biggest reason for the success of cloud computing over the past decade?

The business model has shared success built into it. Customers are renting software rather than buying it — vendors have to make them successful.

What will be the biggest challenge for cloud computing over the coming decade?

The increasing fragmentation and hyper localization of data will make innovation and integration more difficult and more expensive.

Around the Cloud

• Microsoft released more details about last week's big cloud outage, blaming a bad code update for somehow making its way into its production environment despite the usual safeguards.
• Meanwhile, Microsoft has tweaked sales incentive programs to reward salespeople who increase Azure revenue, according to Business Insider.
• Arm is all-in on server processors. CEO Simon Segars said of that market, "we're not in it to be a bit player," predicting that Microsoft and Google will follow AWS in releasing their own Arm designs.
• And Nvidia got into the cloud services business with a new videoconferencing service, as it awaits regulatory approval of its Arm acquisition.
• Alibaba is seeing a lot of the same trends that Nvidia cited in support of its Arm strategy, reporting that it's getting closer to posting a profit from its cloud services division amid a shift to AI-oriented demand from its customers.
• Solo.io is an interesting startup in the very crowded and drama-prone service-mesh space, and it just raised a $23 million series B round of funding.
• The monitoring startup Pixie just raised a $9.15 million round, fresh off the public-market success of monitoring companies like Sumo Logic and Datadog.
• Cloudflare is emerging as an election security force now that lots of local and state governments have taken it up on its free security service, which bears revisiting in a month.
• The Big Tech antitrust report was unveiled Tuesday, and it takes a look at AWS as part of its analysis of Amazon. While it falls for several of the facile "open-core" startup arguments about AWS the cloud sections are still worth a read.
• You really haven't lived until you've listened to Flavor Flav explain Kubernetes. Trust me.

A MESSAGE FROM DIGITAL REALTY

Data Gravity is a Key Megatrend Impacting Enterprise Growth

Digital Realty's new Data Gravity Index DGx™ Report spans industries and geographies. The report helps enterprises understand the dynamics of Data Gravity and how to effectively turn it into an opportunity. Read it here.