Pass on PaaS Until It’s Ready

It’s still much too early for any definitive insights around PaaS. PaaS as a technology is imminent, but has a long way before we can point to some great use-cases. Any vendor in the space could make a move  which places them in the lead position…it’s just too soon to tell.

If IaaS provides IT and application developers a 1X improvement in productivity in the form of faster access to infrastructure, then PaaS promises to provide 4X improvement through faster access to application services.

The Ideal PaaS

The ideal PaaS platform will provide a complete application platform within a multitenant cloud environment (whether private or public or in a hybrid configuration). The platform will also include all development tools (IDE), as well as runtime, administration, and management tools/services.

Some example components of PaaS from the subscriber’s view would include:

  • Subscriber portal
  • Subscription ecom I/F
  • User management
  • App provisioning/orchestration
  • Legacy app migration tools
  • SLA reporting
  • Source-code repository I/F

Components of PaaS from the provider’s view:

  • marketing
  • pricing
  • contracting
  • billing
  • PaaS platform ALM
  • SLA management
  • Provisioning
  • IaaS / Multitenant Platform
  • SaaS security framework
  • Application platform/server (DBMS)
  • Integration platform (ESB, MDM, metadata)
  • UI
  • Modeling platform (BPMS, MRMS, CEP, BI)
  • Application components (web services)

The core of what makes a PaaS more than an IaaS are the application services such as:

  • application design
  • application development
  • testing
  • deployment and hosting
  • team collaboration / developer community facilitation.
  • web service integration and marshalling
  • database integration
  • security
  • scalability
  • persistence
  • state management
  • application versioning
  • application instrumentation

PaaS Adoption Inhibitors

Some of the key inhibitors to current PaaS adoption include:

  • Difficulty of migrating legacy applications into the PaaS platform (new programming language / new SQL schema)
  • Fear of lock-in – because there is no interoperability be/ vendors
  • Lack of application services which assist in go-to-market once the application has been deployed
  • Companies needs a complete platform (and if customization is required, is it possible to make changes?)

Private PaaS: Any Dominant Players?

Hard to say who are the top PaaS candidates for private cloud. It may come down to Red Hat or Citrix with an open approach versus VMware with a proprietary approach.

Red Hat’s OpenShift is an incarnation of the Makara technology that it bought it in November, 2010. It comes in three versions — Express, Flex and Power — and can run atop public or private cloud resources.

VMware’s vFabric 5 is an on-premise product designed to run atop VMware’s vSphere hypervisor and virtualization-management software. It consists of the following core components VMware bought relatively recently:

  • Spring tc Server
  • Spring Insight Operations
  • GemFire distributed database
  • RabbitMQ messaging protocol, Hyperic monitoring software
  • Enterprise-grade Apache web server.

VMware’s recent acquisition, WaveMaker, is a startup focused on letting users build Java cloud applications without having to write code. Whereas Spring betters developers’ lives by simplifying the process of writing Java applications, WaveMaker (which is based on Spring) improves business users’ live by eliminating any writing in the first place. Once they create an application using the WaveMaker visual-development tool, the application is deployed to the user’s cloud of choice.

VMware’s release of Cloud Foundry, an open Platform as a Service (PaaS) offering, provides a platform for building, deploying, and running cloud apps using Spring for Java developers, Rails and Sinatra for Ruby developers, Node.js and other JVM frameworks including Grails. It can run as a public service or be deployed on premise.

The Broader PaaS Landscape

Forrester does a nice job of identifying both private and public PaaS offerings, while categorizing them into the above four categories. Let me attempt to explain…

IDE-neutral: This a class of PaaS where the developers will use IDE tools of choice, but usually within the constraint of one or a few programming languages. For example, EngineYard supports Ruby; Heroku supports Ruby, Node.js and Clojure; Cloudbees for Java; etc.

IDE with Cloud Deployment: This means that the PaaS provider does NOT want to change familiar development environments. So this should read “Your IDE with Our Cloud Deployment”. Developers work in Eclipse or Visual Studio and create code on their local machines. When they are ready to deploy, they upload to the PaaS (dotCloud would be a good example of the most recent entrant in this category).

Cloud IDE: This refers to the fact that the IDE lives in the cloud – meaning that the PaaS provider has integrated the development environment into the hosted service. You use their PaaS for both development AND deployment. Their IDEs would be similar in function as Eclipse or Visual Studio (sounds a bit advanced/aggressive to think one could replace familiar IDEs, but for simple applications this may actually accelerate things for some).

PaaS for Business: This refers to the approach where business experts, not coders, are creating the application. I personally doubt this will take off in any significant way for some time. The concept is further advanced than PaaS itself. How does one code an application without coding? The holy grail is a drag and drop interface of “functions” which connect up with simple business rules and configuration, and way you go!

Now the question is whether you are an Enterprise customer developing internal business applications, a new emerging ISV that is deploying a new SaaS or something in between?

If you’re a Microsoft house full of app-dev, you might lean towards Azure. If you’re a scrappy startup, then the last thing you’ll be looking at is Microsoft, but rather a leading programming language like python, ruby, php, or the like. Some are now looking at Scala, Erlang, and others.

Then you have the private versus public offerings. For example, Tibco sells its Silver offering for internal clouds, and does not provide a public cloud service. That being said, it does allow enterprises deploy its Silver solution on Amazon EC2. Therefore, you could create your own “public PaaS” solution to your employees.

Future of PaaS

In summary:

  • The overall market is extremely immature
  • The many PaaS platforms are primarily startup stories in terms of their customer base
  • Salesforce.com (force) is the most mature (as unattractive as it may be to some)
  • Azure is still trying to complement their PaaS with IaaS
  • Enterprises (not startups) pay top attention to force.com, app engine, and azure
  • Many developers will default to Amazon IaaS for the time being leveraging elastic beanstalk (especially startups)
  • Small players appeal to the small shops (Engine Yard, phpfog, dotcloud)

It’s just a matter of time before all public IaaS providers begin to layer in PaaS services for app-dev teams. In addition, private clouds which initially focused on IaaS, will quickly migrate to supporting application teams (where the real time-to-market and company savings can occur).

I think early adopters in “big Enterprise” will begin projects in private PaaS, but only after they have mastered private IaaS. I believe leaders in private PaaS will involved the financial industry (with a large number of custom web applications) where we’ll begin to see private PaaS, and then hybrid solutions leveraging some public IaaS and PaaS, integrated with internal private clouds.

Of course, there’s the scrappy startup community, who will spin things up over a weekend on the latest public PaaS. Most solutions will be deployed on larger players like Amazon, Rackspace, Engine Yard, Google App Engine, Azure. Frankly, if it were me, I’d just deploy in Amazon to get the best of both worlds (IaaS and PaaS)…..unless you really had an emotional or economic connection with the smaller players. But that’s me.

Keep an eye on the smaller players like Cordys, LongJump, Caspio, WorkXpress, dotCloud, and phpfog.

 

 

 

Jim Kaskade

Jim Kaskade is a serial entrepreneur & enterprise software executive of over 36 years. He is the CEO of Conversica, a leader in Augmented Workforce solutions that help clients attract, acquire, and grow end-customers. He most recently successfully exited a PE-backed SaaS company, Janrain, in the digital identity security space. Prior to identity, he led a digital application business of over 7,000 people ($1B). Prior to that he led a big data & analytics business of over 1,000 ($250M). He was the CEO of a Big Data Cloud company ($50M); was an EIR at PARC (the Bell Labs of Silicon Valley) which resulted in a spinout of an AML AI company; led two separate private cloud software startups; founded of one of the most advanced digital video SaaS companies delivering online and wireless solutions to over 10,000 enterprises; and was involved with three semiconductor startups (two of which he founded, one of which he sold). He started his career engineering massively parallel processing datacenter applications. Jim has an Electrical and Computer Science Engineering degree from University of California, Santa Barbara, with an emphasis in semiconductor design and computer science; and an MBA from the University of San Diego with an emphasis in entrepreneurship and finance.