Gartner on cloud contenders: AWS fails to lower its prices, Microsoft 'cannot guarantee capacity', Google has 'devastating' network outages

There's some interesting criticism in Gartner's latest "Magic Quadrant" report on Cloud Infrastructure and Platform services, which covers just seven providers, the only vendors which it considers qualify for the description "hyperscale cloud providers." The research company has countless magic quadrants, which analyse sectors …

[Peter-Waterman1]

AZ's

This - "What’s wrong with Azure? Capacity. Gartner claimed it has “the lowest ratio of availability zones to regions of any vendor in this Magic Quadrant.”

This is exactly why Azure Services keep falling over, they have 90% of their regions without AZ's and every time someone sets off a fire extinguisher or cuts a cable the whole region goes down...even where they do have AZ's only a few Azure services actually use them!!!

[AMBxx]

Re: AZ's

I was recently deploying a new VM to Azure in the UK. Supposedly, there are 2 locations, but only one provided the combination of bits that I need.

I'm a big fan of Azure, but stuff like this needs fixing.

[Anonymous Coward]

Re: AZ's

We have gone full circle to the Windows Azure Blue Screen of death...

[Anonymous Coward]

Storage Costs

"AWS Storage is identified as an area that “has not experienced a price reduction since 2014,” despite greatly reduced prices for storage components."

At least AWS Charge per GB, unlike Azure that goes in GB increments 2, 4, 8, 16, 32, 64, etc...

[AMBxx]

Re: Storage Costs

In Azure, you just pay for what you're using, not what you've asked for (in most cases).

[Anonymous Coward]

Re: Storage Costs

Nope, defiantly priced on the size of the disk and in doubling increments.

"The total cost of Premium SSD Managed Disks depends on the size and number of the disks,"

https://azure.microsoft.com/en-gb/pricing/details/managed-disks/

[Anonymous Coward]

Cloud.

Lol.

I love it when my competitors 'go cloud'. I know they won't be my competitors for long.

[Anonymous Coward]

Re: I know they won't be my competitors for long.

I love it when my competitors "go internal combustion engine". I know they won't be my competitors for long.

I love it when my competitors "go spinning jenny". I know they won't be my competitors for long.

I love it when my competitors "go agriculture". I know they won't be my competitors for long.

I love it when my competitors "go evolving into land animals". I know they won't be my competitors for long.

[Anonymous Coward]

Re: I know they won't be my competitors for long.

And just how is the cloud an “evolution”? I’m asking on behalf of all rational people.

[Anonymous Coward]

Re: And just how is the cloud an “evolution”?

Me trying to manage my IT infrastructure evolving into me not having to think about my IT infrastructure.

No, actually you're a genius, specialisation is overrated! We should all be cooks, developers, doctors, plumbers, electricians and engineers.

[Peter Gathercole]

Re: Cloud. @AC

Why will they not be your competitors? You going out of business because you can't compete!

[Anonymous Coward]

Leaving AWS

AWS is the stickiest substance in the universe.

[NeilPost]

Re: Leaving AWS

Cloud Jizz ??

[Anonymous Coward]

Gartner in the title of the article...

If Gartner is in the title of the article I don't read the article.

I didn't read it but it doesn't stop me from commenting though.

Why does El Reg bother with Gartner articles, if they want random speculation I can do that.

I might as well rely on social media for information.

So El Reg, want to pay me for making up random bullshit to replace the Gartner stuff you peddle?

I could do an article a month for, let's say. 500 quid.

[Joe Montana]

Re: Gartner in the title of the article...

It's not random speculation, it's paid propaganda.

In 2011 gartner "predicted" that windows phone would leap ahead of iOS in market share by 2015...

Back in 2003 they said that windows mobile would dominate the smartphone market.

Microsoft paid gartner a lot of money for this marketing..

There are some people who actually consider gartner reports to be worth something, and this did result in a few companies standardising on microsoft mobile devices for a time based on reading the gartner reports - only to be forced to quickly move to android or ios devices when the devices they were using got dropped.

They don't actually use or test the products/services they write about, information published by gartner is supplied by the vendors themselves - ie it's "best case" marketing material and doesn't reflect real world experience where advertised functionality is almost never as good as the marketing literature claims it to be.

At most what they do, is compare the claimed feature sets of vendors... Only many vendors will exaggerate their claims, they may have features X Y and Z on paper but that doesn't mean you as a potential customer would need or want those features, nor does it mean that they actually perform as expected.

When it comes to choosing products or services, there really is no substitute for actual experience. There are people who have used a product extensively and know its individual strengths and weaknesses. Every product/service has its own strengths and weaknesses, but which set is best for your individual use case can vary massively.

On another note, why should AWS lower prices so long as people are still buying? Their cost of supplying the service may decrease but they have no reason to pass that saving on to the customer. This is how capitalism works, companies will gouge their customers whenever they are able to.

[Snake]

Re: Gartner in the title of the article...

There's really not much of a problem here. We've / techies have been saying for years that "cloud" only equates to "someone else's computers, somewhere" and that still maintains all the inherent issues with technology, just in someone else's (with little accountability) hands).

Cloud is just another tool, not a "Solution!". Sometimes the tool is good for the job, sometimes not, and sometimes it's only one part of a more complex equation.

[Martin M]

Re: Gartner in the title of the article...

Some techies have indeed been saying for years that "cloud" only equates to "someone else's computers, somewhere". But it's only true in the same sense that a house equates to bricks.

If you're talking about manually standing up VMs and storage in a datacenter through an API or web console, maybe. Although too many companies seem to screw up building and running internal clouds that try to do even that.

But really, what is driving people to cloud providers is access to a huge number - Amazon have 160+ - of highly automated services, all integrated into the same logging, monitoring, billing and identity/access infrastructure and very often into each other as well. Container management, ESBs, data warehouses, SDN, API gateways, VDI farms, call centre infrastructure, software development tooling, ML model lifecycle management, virtual HSMs, machine learning based PII classification, scale-out graph database, managed PostgreSQL, mobile app identity management - too many to sensibly enumerate on a single web page.

These - most of which are at least reasonable, some best in class - are all available for distributed teams to use and manage in 60+ datacentres in tens of countries. With largely good documentation and no need to file a ticket and wait for weeks to get going (or months, if someone has dropped the ball on capacity management). And which can be completely reproducibly stood up via a bit of Terraform - subject to appropriate governance, of course.

If you can point me to a corporate IT department that offers anything close to those 160+ services, with a similar experience, I'll concede it's just other people's computers. I suspect you'll struggle, because the cloud providers probably invested more into R&D for just one of those services than your entire IT budget for the last few years. There are massive economies of scale in automation - cottage industries within enterprises will just struggle to compete.

Of course cloud is just a tool though, and maintains many of the inherent issues with technology - I agree with that. It's just it does solve a useful number of those issues.

[Peter-Waterman1]

Re: Gartner in the title of the article...

Techies that say it's just someone else's computer clearly have no idea about what cloud is. If you use Cloud to just run VM's then you are missing the point, by a long way...

[a_yank_lurker]

Re: Gartner in the title of the article...

Gartner is a well known navel gazing outfit (rarely right) that is widely believed by the upper manglement. It is nice to know what vomit they spewing so as to know what turds will be emitted by your local manglement. Consider it a public service announcing idiocies to come.

[CAPS LOCK]

Re: Gartner in the title of the article...

If you want to know why El Rego prints Gartner stuff I suggest you switch off adblock for a period.

[macjules]

Re: Gartner in the title of the article...

Yeah, strange that "IBM is best' for how many years is that now? Could not have anything to do with how much Gartner are paid by vendors I suppose.

[bigtreeman]

Big Blue

In one paragraph you mention IBM supporting for decades and now owning Red Hat,

a mainstay of Linux and ongoing development,

then say “diminishing mind share among developers who perceive IBM as a provider of legacy technologies.”

What future impact will Red Hat have on IBM ?

I didn't notice IBMs share price taking a dive recently.

I agree with foxglove

[Randall Shimizu]

Re: Big Blue

IBM has strongest private cloud solution of any company today. Today IBM offers full cloud encryption with their private cloud solution. IBM's hybrid multi-cloud solution offers more flexibility than any other cloud solution. Adding cloud fail over is a critical offering for enterprise security. Personally I am hoping that IBM offers full encryption for their hybrid multi-cloud offering.

[Anonymous Coward]

Re: Big Blue

...writes an ibm staffer.

[Anonymous Coward]

Re: Big Blue

Given the level of information content in that post, it’s also possible they’ve just forked scigen, badly deployed Watson. It’s difficult to be sure, given some staffers.

[Concrete Gannet]

Re: Big Blue

"Full cloud encryption" implies other clouds don't encrypt something. What?

To the best of my knowledge all the big cloud companies encrypt data at rest and in motion. "Full cloud encryption" isn't a differentiator. What am I missing?

[Anonymous Coward]

Ah, Gartner, I remember them hugely bigging up the future of Windows Phone, although that report seems to have mysteriously gone missing.

[Maelstorm]

With this report...

With this report, the clouds are burning...along with the forests here in California, U.S.A. Maybe the big three should contract with Cal-Fire to douse the flames of their burning servers.

[naive]

Try increase the size of a disk on an Azure Windows server

It is back to the future of Windows 2000, make backup, destroy disk, make a newer larger one and restore :).

[Missing Semicolon]

"IBM Cloud"

Notice it's mostly for lift-and-shift.

That's because it's really SoftLayer, which was getting old when IBM Borged it - and isn't really "cloud" just VM hosting. Last I looked, the capacity management is utterly pants. There was simply no way to find out if you had enough capacity for a set of VMs that were going to be in the same private network until you create them.

And I bet the random-number billing is no better too.

[Anonymous Coward]

How about some new entrants using the unique distributed computing tech?

I recently came across this technology used by scientific organizations like CERN, SETI and many others that use distributed computing to achieve a much better price/performance ratio.

Few companies like Q Blocks are trying to bring this in the commercial space, specially for AI workloads and the cost differential is dramatic like 10x better.

I see this as clearly growing in the future as network speeds and technology further improves.