Why CIOs must pursue ‘eventual symmetry’ for their cloud strategies

Sinclair is CEO and cofounder of Apprenda

The idea that hybrid cloud is the end state of enterprise computing is no longer controversial. Nearly all technologists, IT executives, and analysts subscribe to the idea that public cloud and on-premises computing both have a place in modern enterprise IT strategy.
A hybrid end state isn’t a bridging tactic or a strategic consolation prize, but a desirable outcome. In fact, a strong case could be made that a hybrid model allows for specialized optimization based on use cases – there are many scenarios both now and in the future that may map best to on-premises or public cloud.
There are two primary ways to implement a hybrid end state: asymmetric and symmetric.

1. Asymmetric – In asymmetric orientations, an enterprise consumes public cloud as one endpoint, and builds an on-premises cloud that is a distinctly separate, second endpoint. For example, we could look at the Infrastructure-as-a-Service (IaaS) layer and say that an enterprise could use OpenStack on-premises and AWS in public cloud, and use processes, operations, and a brokering abstraction across the two endpoints to help normalize the consumption of IaaS regardless of what side of the firewall it came from.

In asymmetric hybridity, the technology used on-premises is different than that used in the public cloud, resulting in the need for reconciliation and the need to accept a lossy factor (i.e. the two technologies may have different features and evolutionary paths) since points of differentiation between the two need to be ignored or marginalized to ensure consistency.

2. Symmetric – Symmetric hybridity means that an enterprises on-premises assets and public assets are using the same technology, and that technology reconciles the assets on both sides of the firewall as a single endpoint. An example of this would be a Platform-as-a-Service (PaaS) layer that can be installed on-premises that could use local OSes and OSes from one or more public clouds all under one logical instance of the PaaS.

The PaaS hides the fact that resources are coming from disparate providers and only exposes that fact where appropriate (e.g. at the policy definition level to shape deployments). In this case, the PaaS is the single endpoint where interaction happens, and resources on both sides are used as resource units by the PaaS. Any organizational processes and consumption processes would be ignorant to the idea that a border exists in the resource model.

Pros and cons of symmetric and asymmetric models

Symmetric models guarantee that anyone within the enterprise consuming cloud infrastructure is shielded from the distinction between on- and off-premises resources and capabilities. If the end user  of cloud infrastructure (e.g. a developer or data scientist) is required to acknowledge any asymmetry, they will have to deal with it in their project. This explicit need to deal with a fractured cloud creates an immediate “tax” related to consuming infrastructure and it will generate consumption biases.
For example, if one side of the asymmetric deployment is easier to consume than the other, then an end user will prefer that even if the not-so-easy side is more aligned with the project, and will cause IT itself a number of headaches when it comes to operations related to that project.
It’s important to understand symmetry doesn’t mean the on-premises and public cloud side of a hybrid deployment must be equal. Certainly, workloads may need on-premises or public assets to satisfy certain requirements the other side couldn’t possibly satisfy.
What symmetry guarantees is that a workload that is indifferent to on-premises or public never be exposed to those concepts. Symmetry also ensures a workload with requirements that can only be satisfied by one part of a hybrid cloud or another is never exposed to the technical divide between the clouds. Instead, a workload communicates its preference in the language of requirements.

Eventual symmetry

Asymmetric models might be good starting points or appropriate for certain layers of the infrastructure stack, but they’re not ideal as a final end state. Symmetric models are clearly superior in almost all other aspects.
In response to this, CIOs should pursue a strategy of ‘Eventual Symmetry.’ Eventual symmetry means that any cloud strategy must:

  1. Choose symmetric models over asymmetric models where possible
  2. If asymmetric is the only possible approach, ensure that the implementation lends itself to eventually being replaced by a symmetric model or that processes and technology be used to abstract the asymmetry into a perceived symmetric model

By establishing eventual symmetry as a core cloud strategy pillar, a CIO can guarantee that any disjointedness in their strategy will be resolved. He or she can also ensure consumers of IT resources are abstracted away from details related to on-premises and off-premises.

How hybrid will reshape the entire cloud market

Sinclair Schuller is the CEO and cofounder of Apprenda, a leader in enterprise Platform as a Service.
When the phrase “hybrid cloud” is mentioned, some technologists will tell you it is the eventual end state of cloud computing, while other technologists chuckle. Those that chuckle typically view hybrid as a phrase used by vendors and customers who have no cloud strategy at all. But hybrid is real and here to stay. Not only is it here to stay, but the hybrid cloud will also reshape cloud computing forever.
People today imagine public cloud to be an “amorphous, infinitely scalable computing ether.” They think moving to the cloud rids themselves of the need to detail with computing specificity and that cloud makes you location, risk and model independent. They think enterprises that move to the cloud no longer need to depend on pesky IT departments and deal with risks associated with centralized computing. This perception of computing independence and scale couldn’t be further from the truth.
The promise of cloud is one where anyone who needs compute and storage can get it in an available, as-needed, and robust manner. Cloud computing providers have perfected availability to the point where, even with occasional mass outages, they outperform the service-level agreements (SLAs) of internal IT departments. This does come at a cost, however.
Cloud computing is arguably the largest centralization of technology the world has ever seen and will see. For whatever reason, many people don’t immediately realize that the cloud is centralized, something that should be heavily scrutinized. Possibly because the marketing behind cloud can be vague and lacking a description of a tangible “place.” Don’t be fooled.
When an enterprise selects a cloud vendor, they’re committing to that provider in a meaningful way. As applications are built for or migrated to a cloud, switching cost gets very high. The nature of this market is driven by a network effect where, assuming all else is equal, each prospective customer of a cloud provider (AWS, Microsoft, etc.) benefits by consuming a cloud that has many of customers over one that has fewer since it indicates lower risk and helps drive the economies that make a given cloud attractive.
If we play this future out, we’ll likely see the cloud infrastructure market collapse to just a few massive, global providers. This will partly be driven by success of the individual providers and the consolidation of smaller players who have great technology but simply can’t compete at that scale. Just take a look at the acquisition of Virtustream by EMC just prior to Dell’s acquisition of EMC for a recent example.
A look at recent market share estimates show exactly that, with Amazon, Microsoft, IBM, and Google accounting for 50 percent of the global cloud infrastructure market. One day, these five vendors will likely account for 80 percent of the market. Compare that to the often-criticized banking world, where despite the massive size of today’s banks, the list of banks that hold 50 percent of global deposits is much longer than just five banks. If we applied the same standard to cloud computing, we’d certainly be infuriated and demanding that these “too big to fail” computing providers be broken up.
To be clear, I’m not suggesting that what’s happening is bad or that public cloud is bad, but rather to point out the realistic state of cloud computing and the risk created by centralizing control to just a few providers. Cloud would likely never have succeeded without a few key companies making massive bets. The idea of a truly decentralized, global cloud would likely have been the wrong starting point.
Let’s explore the idea that a global decentralized cloud, or something more decentralized than what we have now, is the likely end state. Breaking up cloud providers isn’t necessary or optimal. Unlike banking, technology is capable of layers of abstraction to mitigate these sorts of centralized risks.
Most large enterprises looking to adopt cloud are making two large considerations in their decision process:

  1. They can’t shut down their entire IT department and replace it with cloud. There are many practical reasons why this is unlikely.
  2. Many are keenly aware of the risks associated with depending on a single vendor for all their cloud computing needs.

The first consideration makes it difficult to adopt a public cloud without at least considering how to reconcile the differences with on-premises, and the second makes it difficult to choose one provider at a level that is incompatible with another provider. The result of centralization in public cloud providers and looking for symmetry between off- and on-premises computing strategies is driving enterprises to explore (and in some cases demand) hybrid capabilities in layers that abstract away infrastructure. In fact, hybrid has transformed to be synonymous with multi-cloud.
Technical layers like enterprise PaaS software platforms and Cloud Management Platforms have evolved to allow for multi-cloud capabilities to cater to a modality where resources are abstract. Over the coming years, we’ll likely see multi-cloud features in these technology layers to lead to a much more decentralized computing model where something like a PaaS layer will fuse resources from public clouds, on-premises infrastructure, and regional infrastructure providers into logical clouds.
At least in the enterprise space, “private clouds” will really be an amalgam of resources and will behave as the single, “amorphous ether” that we tend to assign cloud to begin with. The cloud market will not be one where five vendors control all the compute and customers are at the mercy of the vendors. Instead, cloud will be consumed through multi-cloud layers that will protect customers from inherent centralization risk. The end state is a decentralized model with control points owned by the customer through software – a drastic reshaping to say the least.

Heroku’s new app-development product line is meant for the enterprise

Heroku, the Salesforce-owned company that powers the application-development process of hot startups like Lyft and Upworthy, announced a new product line Thursday called Heroku Enterprise. It’s geared for big companies that want to develop the kind of modern applications seen at startups while providing the type of features that many large enterprises want, including security features and access control.

Essentially, the product line claims that large enterprises can now have it both ways: a way to make the type of applications that are typically derived from an agile-development process (with access to trendy technology like containers and new database services) all while being monitored under the iron fist of the enterprise. Kudos to Heroku if it can pull that off.

With Heroku Enterprise, organizations can supposedly now monitor all their developers, applications and resources under one interface. Companies can keep tabs on what applications are in production, which developers are working on an app and how each app is eating up resources, according to a Heroku blog post detailing the announcement.

From the blog post:
[blockquote person=”Heroku” attribution=”Heroku”]Heroku Enterprise introduces a new kind of application-level access control called a privilege. Privileges strike a balance between fine-grained permissions that are too hard to manage and coarse-grained, all-or-nothing flags that won’t do the job. In this initial release, we are introducing three app level privileges in beta: deploy, operate and manage. [/blockquote]

The new product line also comes packed with Heroku Connect, which can link up a company’s Salesforce data to the Heroku platform. [company]Salesforce[/company] said that pricing for Heroku Enterprise will be based on how many resources a company consumes.

Of course, developing the types of applications seen at Lyft and Instacart requires a type of developer mindset that can contrast with the old waterfall-style of development seen at big enterprises in which releases don’t come as often and the development lifecycle at large is more sequential in nature.

Even with a new product, it’s important for companies to realize that development is not just tool-centric, but also requires a bit of a culture shift.

Former cloud pariah Oracle claims stronger cloud sales

Oracle, which ramped up its cloud marketing and product rollouts over the past year, touted some encouraging signs for that business in its second quarter, ending November 30.

Revenue from [company]Oracle[/company] cloud products — which fall into what the Wall Street Journal called a “catchall category” — was up 45 percent year over year to $516 million. (Total Oracle revenue was up two percent to about $9.6 billion, from about $9.3 billion last year.)

Revenue from SaaS and PaaS sub-segments of cloud were $361 million, up 41 percent year over year, while IaaS revenue was $155 million, up 62 percent, CEO Safra Catz said on the company’s earnings call Wednesday night. (SeekingAlpha has the transcript.)

And, as usual, you can get a glimpse into what rivals [company]Oracle[/company] is most worried about by the comparisons company execs threw out. Said Catz, for example:

“Overall our cloud results were better than expected as we are clearly growing faster than [company]Salesforce.com [/company]and were more than three times the size of [company]Workday[/company].”

This was Oracle’s first earnings call since company founder Larry Ellison stepped down as CEO in September, ceding that slot to Catz and Mark Hurd.

Oracle 2Q FY 2015 earnings

Ellison, who is now chairman and CTO, sees more good things ahead:

“In Q2 we booked more than $170 million in new SaaS and PaaS annually recurring revenue or ARR. In other words, we sold over $170 million of new SaaS and PaaS annual subscriptions this past quarter.

In Q4 of this fiscal year, we expect to sell more than $250 million of new annual SaaS and PaaS subscriptions. That means, during our next fiscal year we will sell well over $1 billion of new SaaS and PaaS annual subscriptions.”

Later in the call, Ellison said the company expects to see “well in excess of $1 billion in new annual subscriptions … which is about what Salesforce[.com] will be selling in their next fiscal year. I think they are at $1.1 [billion] or something like that, best as we can estimate.”

Nomura Securities analyst Rick Sherlund was cautiously optimistic in a research note, which pointed out that the company still faces “a long transition period to the cloud.”

About 5 percent of total Oracle revenue currently come from those cloud businesses. He wrote:

“We  view this as an encouraging step along the way, but there are still risks of ongoing execution, margins are lower in the cloud, cash flow may be dampened by the need for higher capex to build out data centers as the cloud business scales up, and on-premises license revenues are likely in secular decline. But with stronger growth potential in the cloud, the risk/reward looks favorable to us”

Oracle’s got products, but are they cloud?

A nagging problem for the company, which is the leader by far in on-premises databases and is a giant in enterprise applications, is that many still don’t see Oracle’s cloud products as real cloud products. And here is why:

Oracle DbaaS price chart

Yes, Oracle offers its DbaaS by the hour, if that’s the way you want to purchase it. But, once you hit that “Buy Now” button, you have to make a phone call before you can set up an account. That doesn’t seem very self-service-y.

oracle dbaas slide2

Once the account is set up, presumably things get easier. Maybe this is a distinction without a difference, but I’m betting that people used to buying cloud resources would be shocked to encounter the screen above.


This story was updated at 11:49 a.m. PST to correct my assertion that Oracle DbaaS is not available by the hour. It is, as documented by the chart inserted above.

IBM builds up its cloud with Netezza as a service and NoSQL as software

IBM announced a new, promising collection of cloud data services on Monday, adding to an already-impressive collections services on its Bluemix platform. At this point, though, IBM’s biggest challenge isn’t selling enterprise users on the cloud, but convincing them it’s still the best choice.

Pivotal nabs OpenStack co-creator and Piston co-founder Joshua McKenty

Joshua McKenty, one of the early architects of OpenStack while at NASA, and a co-founder of OpenStack startup Piston, has joined Pivotal at field CTO for Cloud Foundry. He hopes to make Cloud Foundry, running on OpenStack, into what NASA envisioned several years ago.