ARCHITECHT Daily: It's earnings and IPO season for cloud and Cloudera

Earnings season is upon is, and if you've been looking for a reason to suggest that Microsoft Azure m
ARCHITECHT Daily: It's earnings and IPO season for cloud and Cloudera
By ARCHITECHT • Issue #65
Earnings season is upon is, and if you’ve been looking for a reason to suggest that Microsoft Azure might give Amazon Web Services a run for the money, you might finally have a financial one. CNBC has a good, brief rundown of the big three cloud players, but here’s the gist:
  • AWS brought in $3.66 billion in revenue, which was up 42 percent from last year. However, year-over-year growth dropped from last year’s first quarter.
  • Microsoft’s “Intelligent Cloud” unit, which includes Azure, grew 11 percent, to $6.8 billion. Microsoft doesn’t break out Azure revenue specifically, but said Azure saw a 93 percent increase in revenue over last year.
  • Google Cloud is buried somewhere in “Other Bets” on Alphabet earnings, a segment that grew 50 percent to $3.1 billion. 
Here’s a little more info on AWS and Microsoft. If their respective declines and increases in cloud-revenue growth continue for a year, we’ll really have something to talk about.
Intel had a record quarter in terms of revenue, but missed expectations. Its Data Center Group saw revenue increase 6 percent to $4.2 billion.
Of course, the other big financial news on Thursday was that Cloudera priced its IPO at $15 per share, giving it a target raise of $225 million and a target valuation of $1.92 billion. (It’s trading at $18.85 as I type this.) A lot has been made about Cloudera not living up to its $4 billion valuation post-Intel-funding a few years ago, but there’s a counter-argument that people should judge the IPO against Cloudera’s pre-Intel valuation of $1.8 billion, because Intel did not invest as a moneymaking venture. Here’s a good story explaining that rationale, and breaking down the competitive landscape.
Either way, it’s trading now and we’ll get to talk about this every quarter.

Sponsor: Cloudera
Sponsor: Cloudera
Artificial intelligence
This actually sounds like a productive application of machine learning, as it can automate actions across any (or at least many) applications.
Some cold water on Facebook’s bold claims during F8. A lot of experts, not surprisingly, don’t think we’ll have usable consumer brain interfaces or otherwise anytime soon.
… here’s a take on what new legal rights might be necessary for governing what we can do to our brains and other people’s brains. For the time being, I’m more concerned with the law around Amazon Echo and Google Home.
I would note, though, that FDA approval isn’t the same as immunity from lawsuits. The question of why you’d use a system you don’t understand could be a tough one to answer in certain scenarios.
But then you have stories like this one, where you use the tools at your disposal to help people who might not normally have access to screenings for conditions such as cervical cancer.
Chinese investor Kai-Fu Lee says AI will replace half of all jobs in the next decade. Extreme job displacement predictions are the new existential threat predictions.
By and large, at least. This could be troubling, but the open access to info on deep learning and other techniques could help mitigate a lot of patent suits in the space.
Seems like some level of question-answering could be automated by AI, but probably outside of school. Teachers are still hugely important.
This is a clever idea: Test the intelligence of AI by seeing if it can figure out solutions to problems using only what’s available. Knowing all the uses of duct tape would be a feat.  •  Share
Cloud and infrastructure
It’s flexible, scalable and on-demand … and likely a few years from any meaningful mainstream penetration.
This column hits on some of the themes I’ve been covering here (and on the ARCHITECHT site) over the past few months. The real interesting part is Google’s thought/research leadership vs. AWS’s product mastery.
There’s a lot of talk now about the rebirth of PaaS as the thing companies want, with Cloud Foundry being the shining example. The Heroku folks know a thing or two about that space.  •  Share
Speaking of Heroku, it apparently doesn’t scale too well as your company grows. At least according to this writeup by a company called AdStage. I’m actually surprised Docker isn’t mentioned in here.  •  Share
This is a good case study/DockerCon presentation from Visa, which is a paying Docker customer, it turns out. 
This post from the CNCF focuses on the technical challenges Kubernetes users have highlighted, but pay attention to the one about vendor choice, too. Perhaps Heptio will be the answer, but it’s still young.  •  Share
It’s good to see state and local governments getting innovative on tech in ways that the federal government can’t/won’t. Blockchain has huge promise for things where auditability matters.  •  Share
The company is painting a pretty picture, but the LinkedIn consolidation that dinged quarterly revenue might be a bad omen. Equinix will need to keep the customers it inherits from doing the same and moving to the cloud.
All things data
It’s still making more money than all the Hadoop players combined, but it’s also losing steam—very possibly because of them (and the cloud).
It’s a Hortonworks case study, but sheds some light on how Zulily does big data. The stack consists of Google Compute Engine, Google BigQuery, HDP and Tableau.
Did you enjoy this issue?
The most interesting news, analysis, blog posts and research in cloud computing, artificial intelligence and software engineering. Delivered daily to your inbox. Curated by Derrick Harris. Check out the Architecht site at
Carefully curated by ARCHITECHT with Revue. If you were forwarded this newsletter and you like it, you can subscribe here. If you don't want these updates anymore, please unsubscribe here.