Google Cloud CEO Thomas Kurian (Google Photo)
Welcome to Mostly Cloudy, your semi-regular roundup of all the important cloud tech news you need to know! In today’s episode, we’ll try to figure out if any company considering moving their business to the cloud will want to bet on a cloud provider with a countdown clock.
This Week in Cloud: Google Cloud is on the clock
As it will remind you at every opportunity, Google is not a conventional company. It has built a world-class array of technology assets that could theoretically deliver a huge amount of value to the enterprise tech world, but to this day, Google’s top leaders seem to find this world way uncool.
The Information published a fascinating report Tuesday about Google’s long-term plans for its cloud division, which has languished behind Amazon Web Services and Microsoft for years despite the fact that Google practically wrote the blueprint for operating a global distributed computing network. According to the report, Google leaders considered shutting down the operation last year, frustrated with its progress, but agreed to a five-year $20 billion funding plan with the goal of unseating either AWS or Microsoft from the top two market share positions by 2023.
Assuming this is true (The Information has a decent track record), it’s yet another sign that Google’s storied engineering culture is ambivalent at best about working with enterprise tech customers. That it would have even considered pulling the plug on its existing customers in early 2018, a year in which the corporation continued to stack piles of money generated by its advertising cash machine, is somewhat astonishing: what was it planning to tell customers if it made that decision?
That makes this a very damaging leak for Google Cloud. As a betting man, I’ll wager a year’s subscription to Mostly Cloudy that every cloud sales representative at AWS and Microsoft will receive an email this week outlining a few talking points based on The Information’s article to use the next time they find themselves competing against Google Cloud for a contract.
Enterprise tech buyers trying to figure out how to navigate this once-in-a-generation shift in IT want to know that they’re not going to have to do it all over again in a few years. They’re facing pressure from their bosses, who watch football on Sunday afternoons and fire off emails declaring “we need a cloud strategy,” and that makes betting on the indifferent third-place vendor a little harder to justify.
Expect Google Cloud CEO Thomas Kurian and his expanded sales force to be very busy over the next month, reassuring current and prospective customers about its commitment to cloud services. There’s an argument to be made that deadlines help companies focus: Google has struggled to connect with enterprise tech customers, which is a large reason why Kurian, a 20-year veteran of Oracle’s enterprise sales culture, is now running the show.
Alphabet CEO Sundar Pichai speaks at Google Cloud Next 2019 (Google Photo)
But 2023 is not that far away, and catching either AWS or Microsoft will be an enormous challenge for Google.
AWS is not going anywhere in the next three years, light-years ahead of Google Cloud on revenue and the default choice for enterprise tech buyers who aren’t sure what to do. AWS probably signed multiple three-year cloud deals in 2019 that will generate more revenue than Google Cloud has earned over the last three years, assuming Google’s infrastructure cloud business is doing around $1 billion a quarter, as the report suggests.
Microsoft is a closer target, but Microsoft’s enterprise sales culture dates back into the 1990s; this is a company that knows what makes enterprise buyers tick. It is swiftly converting older customers running Windows Server and other data-center software products to the Azure cloud, and it is the natural fallback option for companies that have no interest in funding Amazon’s retail empire.
That makes for a very narrow path if Google is going to reach its goal. Now that Google co-founders Larry Page and Sergey Brin — the historical roadblocks toward Google’s enterprise ambitions — have stepped down from day-to-day positions at Alphabet, new CEO Sundar Pichai might have more room to push Google’s cloud ambitions.
But the article rightly compares the current state of the market to the state of internet search a decade ago, when Google was the company that ran away with the market despite prolonged challenges from Microsoft and Yahoo. I made a similar argument last year, right around the time Google was apparently thinking about pulling the plug on its cloud business.
The barriers to entry in cloud computing are ridiculously high; there are only a handful of companies on the planet that have the existing infrastructure and the resources to make future investments in that infrastructure, and Google is one of them. This market needs competition, and it would be a shame to see Google give up on its cloud business without having ever really embraced it.
But here’s one for the prediction files, if Google gives up on the cloud: assuming it hasn’t been blasted into the sun by the Justice Department by 2023, Facebook will enter the cloud services market.
And Now, A Word…
Prominent media organizations should spend more time and resources covering cloud and enterprise tech, because on the rare occasions they attempt to cover this space, what you tend to find is something along the lines of what The New York Times published on Sunday about AWS. I had some pointed thoughts about that story, which was based on a profound misunderstanding of how cloud technology, open-source software, and competitive pressures work in this industry.
AWS deserves everyone’s scrutiny, because it has the potential to wield immense power over the entire tech economy for decades to come. But proper criticism must be based on an understanding of the enterprise market and the technologies that make it such a force, otherwise that scrutiny is easily dismissed by companies that would love to believe the media is out to get them.
Around the Cloud
AWS Outposts by the Numbers: A Far-Too-Deep Dive Into Pricing (Adam Leventhal)
An AWS Outposts rack on the show floor at re:Invent 2019 (Mostly Cloudy Photo)
Details about Outposts have been hard to come by, even after AWS formally released the cloud-in-a-box server rack earlier this month at re:Invent. Anyone interested in using Outposts as part of a hybrid cloud strategy might want to check out this post, which breaks down the various configurations available as well as how much they’ll cost you.
An Amazon cloud vice president explains how AWS is taking on Microsoft on its own turf (Business Insider)
Speaking of Outposts, hybrid cloud hardware was not invented at AWS. Microsoft has been selling a similar product called Azure Stack for a few years now, and expect the sales rhetoric to heat up now that multiple options are available in the market.
Stratoscale closes down, lays off 60 (Globes)
Stratoscale was onto this type of hybrid cloud product early, but the small Israeli company was unable to translate its well-received products into revenue, especially now that all the big players are on board. CEO Ariel Maislos said that he expects “traditional players” will bid on its technology as cloud late-comers look for options from their current vendors.
8 data center predictions for 2020 (Tech Republic)
It’s the most wonderful time of the year, when tech news slows to a crawl and everyone compiles “best of the year/decade” lists and predictions for the year to come. These bets from Tech Republic are based on solid ground, although I personally think edge computing is going to evolve a little more slowly.
A Sobering Message About the Future at AI's Biggest Party (Wired)
It looks like AI Winter is coming, yet again. For decades, sentiment around artificial intelligence research has swung between short-term optimism and long-term reality, and the best minds in AI appear to be retrenching as the robots, autonomous cars, and medical breakthroughs we’ve been promised continue to slip into the future.
Google details its approach to cloud-native security (Techcrunch)
Security is always a good bet for cloud providers trying to attract new business, and Google’s internal security practices are a great selling point for its cloud customers. BeyondProd builds upon the idea of “zero-trust” networks that secure microservices architectures across a distributed cloud network.
Google Cloud partners with Fortinet, McAfee, Palo Alto and other vendors to bolster its security (Fierce Telecom)
BeyondProd is probably a little too much for companies that are still running old-school tech infrastructure, which means Google needs to meet those customers on their own terms. New partnerships with traditional security vendors should help get its cloud technology in front of enterprise customers concerned about cloud security.
Diabetes Blood-Sugar Data Outage Blamed on Cloud Provider Switch (The Wall Street Journal)
This headline is a little unfair, given that DexCom CEO Kevin Sayer acknowledged that the company’s internet-connected blood-sugar monitors were not “properly configured” for the new cloud environment. It sounds like Microsoft and Google are the two cloud providers involved, but it’s not clear which company won the new business.
LogMeIn To Be Bought By Two Private Equity Firms For $4.3 Billion (CRN)
LogMeIn was a bit ahead of its time in the single sign-on market, which helps companies allow their workforce to access sensitive corporate information without having to verify their authenticity again and again. Taking money from private equity is always a dance with the devil, but it has worked out pretty well for some enterprise software companies over the last few years.
Former Nintendo Exec Reggie Fils-Aimé Thinks Cloud Gaming Will Bring Everyone Together (ScreenRant)
Cloud gaming makes so much sense for the same reason cloud computing makes so much sense: why manage software updates and new program installation on your own hardware? There are still lots of issues to be worked out, as the troubled launch of Google Stadia shows, but this is an entertainment model that cloud providers will figure out over time.
Majority of financial professionals don’t trust cloud computing (Financial Director)
Well, the majority of the rest of us don’t trust financial professionals, so there you go. Financial companies used to be some of the earliest adopters of new enterprise technology, but legal and cultural hangups have delayed their entry on to the cloud, with some notable exceptions.
Power of the cloud builders: Who ate one-third of the $38bn data centre pie in Q3? AWS, Microsoft, Google 'n' pals (The Register)
The latest market share numbers from Synergy Research underscore the reality of the market for traditional IT hardware in the cloud era: spending by the big cloud companies on new gear “has increased threefold” over the last five years, while enterprise data center operators spent just six percent more over that period. Adding insult to injury, most “hyperscale” cloud providers like AWS and the lot build their own servers rather than buying off the shelf gear.
How Kubernetes Has Changed The Face Of Hybrid Cloud (Forbes)
Kubernetes has immense potential to act as the unifying bridge between multiple public cloud and on-premises data centers, which accounts for a lot of the hype. Forbes takes a look at how Kubernetes can be the common layer for companies that want to indulge in some of the benefits of the public cloud but can’t or won’t move everything out of their own data centers.