Le cloudvupardesexperts 9pov-curationparloicsimon-clubclouddespartenaires


Published on

9 points de vue d'experts sur le cloud. Sélection d'articles issus de la veille et de la curation faire par Loic Simon pour le Compte des membres du Club Cloud des Partenaires

Published in: Business, Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Le cloudvupardesexperts 9pov-curationparloicsimon-clubclouddespartenaires

  1. 1. Voici un peu de lecture pour l’été avec 9 articles issus de la veille/curation Cloud que je réalise quasi quotidiennement pour lesmembres du Club Cloud des Partenaires.Retrouvez également sur le Blog du Club Cloud des Partenaires 2 autres ebooks quiprésentent mon propre point de vue sur l’écosystème Cloud et sur l’intérêt des blogs pourvotre Marketing.Sommaire ● Devops and the Cloud ● Devops is a Verb ● I’ve virtualized my systems…isn’t that a cloud? ● End-to-End Cloud Offerings for Large Enterprises ● NoOps Is As Legitimate As DevOps ● IBM Research Shows How the Cloud is Driving Business Model Innovation ● Cloud Computing Goes Far Beyond Virtualization ● Top 5 Things The Cloud Is Not ● CIOs Dont Need to Be Business LeadersDevOps and the Cloud
  2. 2. June 18, 2012 2:05 pm by Edwin SchoutenOne of the hot topics around these days in IT is DevOps. But what is it exactly, why would Iwant it, and if so, how do I get it? This blog posting discusses these three questions; it’s upto you to decide when you start using it.What is it?Are you, as developers, fed up with theoperations group that doesn’t understand yourcoding, while taking all the time in the world tomake a mess of implementing your application?Or are you from operations, frustrated thatyou’ve been given a half-developed, half-testedapplication that needs to be implemented withoutproper documentation? Or are you from the business side, overlooking the strugglebetween development and operations while wondering why it all takes so long to get this not-too-difficult request implemented? For all of you; DevOps is here!The name DevOps is derived from a combination of the two words development andoperations. DevOps is more than a new development methodology like agile softwaredevelopment, it’s about communication and collaboration between all three statedstakeholders (development, operations, and business) within an organization. It is mainlytargeted at product delivery, quality testing, feature development, and maintenance releasesin order to improve reliability, security, and faster development and deployment cycles.To support DevOps, collaborative tools are needed to support the agile service deliveryapproach, accelerating application deployment from weeks to minutes.DevOps has received increased attention over the last year or so, which makes perfectsense for two reasons: the application landscapes are becoming increasingly complex, thetime-to-market of new functionality needs to be decreased. Organizations need to reducecost while maintaining a satisfactory level of quality — is DevOps a solution to this problem?Why would I want it?An IBM CIO study of hundreds of companies revealed that a number of organizationsare struggling to just get their software into production consistently. In fact, 50 percent ofdeployed applications must be rolled back, with rework accounting for more than 30 percentof project costs. Ultimately, the driver is to reduce the costs of managing applications whilekeeping agile to be able to quickly respond to market demand.As Werner Vogels, the CTO of Amazon, explains in his presentation in 2011 for HackFwdvideo (at minutes 3:00 – 5:00), Amazon struggled with the exact same problem: an immenseunmanageable application landscape. The solution as he explains it is that each service (setof functionalities) is developed and operated by a small team that can be no larger than itcan be fed on two pizzas. Even shorter is “you build it, you run it.”Now I’m not saying that every organization should do the exact same, but the underlyingthought is DevOps. Make functionality and maintainability a shared developers/operationsresponsibility, supported by focusing on inter-team collaboration and communication.As Ovum, an organization that provides clients with independent and objective analysis,describes it in its article: “The solution is to provide both teams with a shared objectivethat is described in business outcomes. This comes from a governance layer that mustmandate the behaviours. The roadmaps and demos shown at Pulse indicate that IBMclearly “gets” this and is working to bridge the gap between development and operations atall levels”.
  3. 3. How do I start using it?Back in February 2011 an excellent white paper Collaborative DevOps with Rational andTivoli was made available. It described the challenges that exist between developmentand operations. It also described how integrations between products from IBM Rationaland IBM Tivoli support effective collaboration to achieve improved accuracy, efficiency,agility, and security in the deployment and monitoring of software systems. The scope ofthe paper spans the areas of strategic planning, deployment planning, automation, and theidentification and remediation of production problems.Sure you can implement DevOps tools yourself in your owndata center, but wouldn’t it be great to get this “as a service”from the cloud? No hardware, installation, or licenses to worryabout. No long-term investments needed. Just switch it on andstart using it. This is now possible using IBM SmartCloudApplication Services. At the platform as a service web page,click the picture (4:08) shown on the right [i1] to obtain a quickbut thorough understanding of how DevOps can work for your business as offered fromthe IBM SmartCloud.And what’s even better; you can register now for the pilot program for IBM SmartCloudApplication Services to be able to use DevOps from the cloud! Just navigate to IBMSmartCloud and follow the instructions — it’s that simple.Still hungry for more? Go to the IBM SmartCloud Continuous Delivery web page that holds awealth of information about DevOps and the various implementation scenarios.About Edwin SchoutenEdwin is the Cloud Services Leader for IBM Global Technology Services in the Beneluxregion (Belgium, Netherlands and Luxembourg) and IT Architect by hart. Edwin has almost15 years experience in IT, of which the last 8 years in IT Architecture, which is backed-upwith a Masters of Science degree in IT Architecture. He is an optimist by nature, analytic butrealistic and has a can-do mentality. He has an ever growing drive to add business valueusing IT, that’s also where his biggest strength comes in; the ability to communicate withboth business and IT.Devops is a Verbposted on Wednesday, June 20, 2012 4:28 AM#devops Devops is not something you build, it’s something you do
  4. 4. Operations is increasingly responsible for deploying and managing applications within thisarchitecture, requiring traditionally developer-oriented skills like integration, programmingand testing as well as greater collaboration to meet business and operational goals forperformance, security, and availability. To maintain the economy of scale necessary to keepup with the volatility of modern data center environments, operations is adopting moderndevelopment methodologies and practices.cloud computing and virtualization have elevated the API as the next generationmanagement paradigm across IT, driven by the proliferation of virtualization and pressureon IT to become more efficient. In response, infrastructure is becoming more programmable,allowing IT to automate, integrate and manage continuous delivery of applications within thecontext of an overarching operationalframework.The role of infrastructure vendors in devopsis to enable the automation, integration, andlifecycle management of applications andinfrastructure services through APIs,programmable interfaces and reusableservices. By embracing the toolsets, APIs,and methodologies of devops, infrastructurevendors can enable IT to create repeatableprocesses with faster feedback mechanismsthat support the continuous and dynamic delivery cycle required to achieve efficiency andstability within operations.DEVOPS MORE THAN ORCHESTRATING VM PROVISIONINGMost of the attention paid to devops today is focused on automating the virtual machineprovisioning process. Do you use scripts? Cloned images? Boot scripts or APIs? OpenSource tools?
  5. 5. But devops is more than that and it’s not what you use. You don’t suddenly get to claimyou’re “doing devops” because you use a framework instead of custom scripts, or vice-versa. Devops is a broader, iterative agile methodology that enables refinement andeventually optimization of operational processes. Devops is lifecycle management with thegoal of continuous delivery of applications achieved through the discovery, refinement andoptimization of repeatable processes. Those processes must necessarily extend beyond thevirtual machine. The bulk of time required to deploy an application to the end-user lies notin provisioning it, but in provisioning it in the context of the entire application delivery chain.Security, access, web application security, load balancing, acceleration, optimization. Theseare the services that comprise an application delivery network, through which the applicationis secured, optimized and accelerated. These services must be defined and provisioned aswell. Through the iterative development of the appropriate (read: most optimal) policies todeliver specific applications, devops is able to refine the policies and the process until it isrepeatable.Like enterprise architects, devops practitioners will see patterns emerge from the repetitionthat clearly indicate an ability to reuse operational processes and make them repeatable.Codifying in some way these patterns shortens the overall process. Iterations refine untilthe process is optimized and applications can be completely deployed in as short a timeas possible. And like enterprise architects, devops practitioners know that these processesspan the silos that exist in data centers today. From development to security to the network;the process of deploying an application to the end-user requires components from eachof these concerns and thus devops must figure out how to build bridges between the ivorytowers of the data center. Devops must discern how best to integrate processes from eachconcern into a holistic, application-focused operational deployment process.To achieve this, infrastructure must be programmable, it must present the means by whichit can be included the processes. We know, for example, that there are over 1200 networkattributes spanning multiple concerns that must be configured in the application deliverynetwork to successfully deploy Microsoft Exchange to ensure it is secure, fast and available.Codifying that piece of the deployment equation as a repeatable, automated process goesa long way toward reducing the average time to end-user from 3 months down to somethingmore acceptable.Infrastructure vendors must seek to aid those on their devops journey by not only providingthe APIs and programmable interfaces, but actively building an ecosystem of devops-focused solutions that can be delivered to devops practitioners. It is not enough to say “hereis an API”, go forth and integrate. Devops practitioners are not developers, and while an
  6. 6. API in some cases may be exactly what is required, more often than not organizations areadopting platforms and frameworks through which devops will be executed. Infrastructurevendors must recognize this reality and cooperatively develop the integrations and themeans to codify repeatable patterns. The collaboration across silos in the data centeris difficult, but necessary. Infrastructure vendors who cross market lines, as it were, tocooperatively develop integrations that address the technological concerns of collaborationwill make the people and process collaboration responsibility of devops a much less difficulttask.Devops is not something you build, it’s something you do.Lori MacVittie is responsible for education and evangelism of application services availableacross F5’s entire product suite. Her role includes authorship of technical materials andparticipation in a number of community-based forums and industry standards organizations,among other efforts. MacVittie has extensive programming experience as an applicationarchitect, as well as network and systems development and administration expertise. Prior tojoining F5, MacVittie was an award-winning Senior Technology Editor at Network ComputingMagazine, where she conducted product research and evaluation focused on integrationwith application and network architectures, and authored articles on a variety of topicsaimed at IT professionals. Her most recent area of focus included SOA-related products andarchitectures. She holds a B.S. in Information and Computing Science from the Universityof Wisconsin at Green Bay, and an M.S. in Computer Science from Nova SoutheasternUniversity.I’ve virtualized my systems…isn’t thata cloud?April 17, 2012 2:12 pm by Joe BohnWhen many people think of cloud computing they immediately think of virtualization andvirtual machines in particular. This is completely natural and not at all surprising. After all,one of the core underlying technologies used in cloud computing is virtualization. However,it is important not to confuse one element of cloud computing with the entire thing – and thiscan sometimes happen. Actually, I don’t really think people literally confuse virtualizationwith cloud computing – but I have heard people refer to their collection of virtual imagesas their “private cloud.” They are too easily satisfied and view their collection of virtualmachines as being “good enough.” They don’t see how moving to a real cloud – private,public, or hybrid – could transform their data center. They are greatly mistaken.First, let’s consider what is typically meant by cloud computing. I think we need an impartialdefinition here so let’s look at what has been produced by the National Institute of Standardsand Technology (NIST). It’s their job to define standards and guidelines, including minimumrequirements, for use in federal agencies and nongovernmental organizations. NISTpublished this definition of cloud computing in September of 2011. The definition is verycomplete and yet also concise (just two real pages of definition). They define the fiveessential characteristics of the cloud model: three service models and four deploymentmodels.Let’s start by looking at the essential characteristics: ● On-demand, self-service ● Broad network access ● Resource pooling ● Rapid elasticity
  7. 7. ● Measured serviceSo, for a solution to be called a cloud it should meet these essential characteristics. Inthis list, it is clear that managing a collection of virtual machines is certainly not a cloud.You must be able to have resources allocated when they are needed and in a self-servicefashion. You must be able to do this from anywhere with broad network access. You mustprovide resource pooling for use by multiple consumers in a multitenant model based uponconsumer demand. Indeed, virtualization certainly comes into play with resource pooling– but only to the extent that it can be leveraged to easily manage and move workloads.The consumers should not even be concerned with the physical location of the workloads.Elasticity is imperative and must be supported to both scale outward and inward as demanddictates, optimally in an automatic fashion. This is something not easily done when youare just managing virtual machines. Finally, the ability to measure usage of the services isimportant and should provide transparency for both the provider and consumer of the utilizedservice.The NIST also defines the three primary service models: software as a service (SaaS),platform as a service (PaaS), and infrastructure as a service (IaaS); and also four primarydeployment models: private cloud, community cloud, public cloud, and hybrid cloud.So I think this gives us a good working model for cloud computing and the value that it canbring over only virtual machines. That isn’t to say that virtualization isn’t important. I think itis very important for cloud computing – bordering on critical. It’s just that virtualization mustbe thought of within this broader notion of cloud computing to gain all of the value that thisnew paradigm has to offer. We’ve been aware of this for some time at IBM and you caneasily see it in the solutions that we’ve delivered for both private and public clouds. I thinkour competitors are just now starting to see this too and are beginning to think beyond onlyvirtual machines. But there are many differences between our cloud solutions and thoseof our competitors. One element is that our competitors are often focused only on theirparticular hypervisor (virtualization infrastructure) technologies to facilitate cloud solutions;IBM, however, gives you a choice.The next question that I think you should ask is – “what role does virtualization play in thecloud and what exactly is it that I should virtualize?” You can probably guess that it is morethan only virtual machines. At IBM, we have always been taking the broader view – lookingbeyond where we are today so that we are prepared for tomorrow.So what does it mean to take a broader view of virtualization? At its core, virtualizationis about introducing a level of abstraction between the producer and the consumer ofsomething. We began this journey by virtualizing the hardware – memory, CPU, storage,networking, and others. This led to virtual machines combining these building blockstypically with an operating system.Let’s not stop there – we can do so much more as we move up the software stack. Forexample, IBM Workload Deployer and its predecessor the WebSphere Cloudburst Appliancehave provided what we call Virtual System Patterns for three years now. The motivationbehind this is that – although virtual machines are great – there are very few businesssolutions that are only dependent upon a single virtual machine running some software.We saw a need to create an abstraction of a complete system with multiple federatedmachines to support complete application solutions provisioned as a single entity. Weprovide deep integration for complete systems built upon standards to support middlewareprovisioning; leveraging best practices, and based on our years of customer experience.We also provide utilities in an open, extendable structure to support customization andintegration of third-party solutions. This is all accomplished using a patterns-basedapproach with a very simple drag-and-drop interface. Once more, our competitors are juststarting to play catch-up here by introducing similar concepts with graphical user interfacesthat allow you to build topologies of virtual systems. It’s amusing to see how similar they
  8. 8. look to what we’ve had for years. I guess imitation really is the most sincere form of flattery.Once again, we didn’t stop there. We took the abstraction up a level higher and introducedapplication virtualization last year in our IBM Workload Deployer private cloud managementsolution. By application virtualization I mean providing the capabilities to abstract theapplication from the underlying infrastructure such that it can be elastic, highly available, andprovide agility across a pool of application infrastructure resources.This type of application virtualization is built into our virtual application pattern (hence thename) – an application-centric way of defining, provisioning, and managing the completelifecycle of your application. Features such as elasticity of the application itself and sharedservices to support non-functional requirements are delivered in policies using commonmetrics such as response time service level agreements (SLA). Requirements that arecommon for nearly any application, such as high availability, are “baked” right into thesolution without any definition required. Virtual application patterns support specific types ofapplications in a highly integrated solution – integrated both on the front-end user interface,and on the back-end implementation of the running systems. Management is from anapplication perspective – not focused on the various middle components that are necessaryto support the application. This a true platform as a service (PaaS) solution where IBMWorkload Deployer dynamically builds the necessary platform infrastructure to support thespecific needs of the application. I don’t see anything similar in scope and user simplicityfrom our competitors – they’ll be playing catch-up yet again.So this is what cloud computing is all about. It is about much more than simply virtualization– it is about transforming the data center. It is about innovation and simplicity. IBM isincluding intelligence into the solutions that we provide and integrating the expertise we havegained over years of experience right into those solutions to simplify IT for our customers.We are providing our customers with the IT for their business so that they don’t have to be inthe business of IT themselves.
  9. 9. But we’re not done yet! You’ve no doubt heard a lot of buzz about our recently announcedexpert integrated systems – and in particular IBM PureApplication System – a system withbuilt-in expertise, integrated by design, and a simplified experience. As you can see we’vebeen on this mission for a while and we’re continuing to build our expertise and knowledgeinto these systems to simplify IT. IBM PureSystems is here! To find out more visit http://ibm.co/HxXzwB.Joe Bohn is a senior software architect at IBM. He is currently serving as a technicalevangelist for WebSphere cloud computing. Before becoming a technical evangelist, herepresented IBM on OSGi Alliance specifications and open source projects includingApache Aries and Apache Geronimo. Prior to these activities he worked in multiple productareas within IBM as architect, designer, team lead, and developer on a variety of IBMproducts including the Integrated Solutions Console, Tivoli Presentation Services, andmultiple Tivoli solutions.End-to-End Cloud Offerings for LargeEnterprisesCloud Computing JournalBy Srinivasan Sundara RajanWith cloud adoption becoming a de-facto option for small and medium enterprises, largeenterprises are relatively slower in their adoption of cloud. The main reason is that largeenterprises have a very complex existing IT setup and no single offering from various cloudproviders has yet to satisfy all their needs.However we find the recent announcements and offerings from IBM provide a perfectplatform for large enterprises to on board to Cloud to make their businesses more agile.Blueprint of Large Enterprises on Cloud-Enabled ITThe following reference architecture provides an overview of how the large enterpriseswould like to position their cloud-enabled IT so that they get best of their traditionaloperations while merging with the benefits of cloud.
  10. 10. Mission-Critical WorkloadsThese will be bread and butter for any large enterprise IT. For example if the enterprise is inthe telecom domain, the workloads like provisioning, billing and network management willbe a part of this workload category. If the enterprise is from the manufacturing domain, thenproduct life cycle management, warranty analysis, supply chain management will be a partof this.Large enterprises utmost concern is the proper functioning of these workloads. Typicallyenterprises will receive the maximum benefit if these workloads are moved to privateclouds which can handle heterogeneous platforms like mainframe, Unix flavors, Linux andWindows. Typically enterprises would like this workload to run without any changes evenafter migration to cloud.Elastic WorkloadsThis is the new terminology coined to represent the workloads, which may not be 100%mission-critical for the enterprises, but are still important, but whose processing is dynamicin nature and are the ones that benefit from the elastic nature of the public clouds.Examples like data warehousing systems, content management systems, massively parallelprocessing operations, consumer-facing web sites form part of this workload. This workloadwill receive the highest benefit for migrating into public clouds.Consumable WorkloadsWith the great proliferation of SaaS (Software as a Service), enterprises have many avenuesto consume certain types of workloads directly from a public cloud without even hostingthem. Popular CRM software like Salesforce are good examples. There are so many otherBpaaS players and situations that I have been covering separately in my ‘Industry SaaS
  11. 11. Series form part of this workload.Cloud IntegrationWith the large enterprises running their business partly in on-premise private clouds andpartly on public clouds, robust application integration options are required to ensure thatthe applications are in sync with each other and information generated out of the extendedenterprise is consistent and preserves the integrity across transactions.Hybrid Environment ManagementWith the increased workload management across the public clouds and the new ways ofmonitoring the virtualization techniques adopted by private clouds, enterprises needed tomove away from the traditional ways of IT environment management and move towards ahybrid environment management.Application Development Life Cycle for the Cloud EnvironmentNo enterprise can afford to be static with respect to their business capability needs. Thismeans the cloud-enabled IT has to rapidly expand to the new market needs. This is onlypossible if the application development platform is cloud-aware and can support both privatecloud deployment and public cloud deployment. The platform should also free enterprisesfrom future upgrades and maintenance activities so that enterprises can become truly agile.PaaS (Platform as a Service) is the best option for large enterprises in this regard.Enterprise Innovation with CloudWhile the above mentioned components of the reference architecture are still part of thetraditional enterprises even without cloud in a different form, however in todays competitiveenvironment large enterprises needed some options which are truly innovative anddisruptive in nature. Cloud value proposition like high performance computing (HPC) andsome other research options form part of this.IBM SmartCloud Product Mapping for the BlueprintSmartCloud is the IBM vision for cloud computing, which accelerates businesstransformation with capabilities from IBM cloud offerings. The following product mapping willhelp to realize the blue print for cloud enablement for large enterprises with the respectiveSmartCloud offerings.Mission-Critical WorkloadsAs explained large enterprises will choose private clouds for running their mission criticalworkloads. IBM SmartCloud has a multitude of private cloud infrastructure and associatedsoftware options. ● IBM SmartCloud Entry on Power Systems. It works with a clients existing Power Systems infrastructure. ● IBM zEnterprise heterogeneous cloud solution ● IBMStarter Kit for Cloudx86 EditionAdditionally there are IBM SmartCloud Foundation infrastructure offerings that includeservers, storage and virtualization components for building private and hybrid cloudcomputing environment.Together the above options help an enterprise to run their mission critical, Mainframe, BigEndian Unix, Linux and Windows workloads on a private cloud.Elastic WorkloadsPublic Cloud offerings form part of the management of these kinds of workloads. IBMSmartCloud Enterprise and Enterprise+ provide the enterprise class public cloud platforms,with the following major features. ● Greater choice and flexibility with Enterprise class operating systems and software images
  12. 12. ● Availability and performance ● Security and isolation ● Payment and billingConsumable WorkloadsSaaS and BpaaS offerings form part of this workload, and IBM SmartCloud has a varietyof SaaS offerings, and like any other community contributed application exchange this cangrow further in the future. Current offerings are categorized into: ● Business Process Management ● Analytics ● Social Business ● Government Offerings ● Buying, Procurement and Sourcing ● Selling and Merchandising ● Marketing and Web Analytics ● Other Business Process As A Service ● Payment and billingTo facilitate a collaborative ecosystem for more SaaS/ BpaaS products ‘IBM ApplicationDevelopment Services For Cloud help partners to develop new SaaS applications.Cloud-Based IntegrationContinuous functioning of large enterprises on hybrid environments (i.e., public, privateclouds) is only possible if there are strong cloud integration platforms. WebSphere Cast IronCloud Integration is part of SmartCloud platform and enables cloud application to applicationintegration.Hybrid Environment ManagementIBM SmartCloud Monitoring is a cloud monitoring tool for cloud infrastructure as well as thevirtual servers running within it. SmartCloud Monitoring is designed to monitor very largeenvironments. Coverage for KVM, VMware, Citrix XenServer, Citrix XenDesktop CitrixXenApp, Cisco UCS, NetApp.With respect to Public Cloud, IBM SmartCloud Enterprise+ provides robust managementoptions.Application Development Life Cycle for Cloud EnvironmentsIBM SmartCloud Application Services allow your organization to develop, deploy, manageand integrate applications in the cloud. Initially, IBM SmartCloud Application Services willprovide initial support for Java, and expand to include PHP, Ruby, C, C++, .Net and otherslater. IBM SmartCloud Application Services portal, tooling will be provided to manage thedeployment and management of applications. In addition, Rational products will be availablefor application lifecycle management. The service will use the IBM DB2 Enterprise Edition9.7 database, which is Oracle-compatible. Data can be extracted from an Oracle databaseand moved into the DB2 database.Enterprise Innovation with CloudThe HPC cloud offerings from IBM provide the methods and means to manage HPCinfrastructure using cloud computing technology. The HPC cloud offerings from IBMcomplement the IBM SmartCloud.SummaryLot of material is available on the IBM site about the detailed offerings as part of IBMSmartCloud. The aim of this article is to position how the large enterprises can adopt to thisrobust cloud offering and can take comfort of an end-to-end offering, which will ensure thatthey handle any fear or uncertainty or doubt among the stake holders.
  13. 13. NoOps Is As Legitimate As DevOpsBy Krishnan Subramanian on March 14, 2012Ever since Lucas Carlson, CEO of AppFog, brought the term “NoOps” into the focus ofdiscussion, there is quite a bit of backlash against the term. The debate sometimes bordersalong insanity and I thought I will add my 2 cents to this cacophony. In fact, this backlashis nothing new. Whenever I make a statement about the role of ops fading away in a cloudbased world, I get similar brickbats on Twitter and other online fora. Let me use this post toadd clarity to the point I am advocating with respect to Ops. Before the Ops guys and galspounce on me, I also want to highlight that I am not a developer and my background is onthe ops side. Having made it clear, let me add my thoughts on why NoOps is a legitimateuse of the term, much like DevOps.It’s all fine, what is NoOps, BTW?Regular readers of my blog know that I am bullish on PaaS being the future of cloudservices. Right now, PaaS adoption is in the early adopter stage and in the next five years,it is expected to be mainstream in the enterprise. When that happens and enterprisesadopt hosted PaaS from providers like Heroku, Engine Yard, CloudBees, AppFog, Azure,etc., there will be no need for enterprises to invest on operations as these PaaS offeringsgives an interface to their developers which they can use to build and deploy their appswithout worrying about the underlying infrastructure including security and scaling. Thisscenario is what Lucas calls as NoOps and I have also emphasized it on my posts with theslogan “Forget DevOps, embrace the damn PaaS”.So? What is the big deal?When such a transformation happens in the industry, the role of operations people is goingto be diminished compared to what it is today. Many ops people and pundits (with theirhearts on the operations side) take it personally and argue that NoOps is bordering on FUDand operations are not going away anytime. They push back against the term because itseems to suggest that operations are going to vanish in the coming years. Their argument isthat operations are critical part of these technologies including PaaS and SaaS and any termthat diminishes their value is just a marketing term with FUD value.You are part of the marketing FUD, Huh?Not really. Whether we (pundits and ops people) like it or not, even the public cloud serviceslike AWS made ops people less visible. Gone or the days when a developer will put a helpdesk ticket and wait for IT to provision a server for his/her needs. Today, the self servicepart of the cloud offerings lets developers provision the instances needed for them with afew clicks or an API call. Since public clouds offers them a way to operate the infrastructurethrough code, the DevOps movement came into picture calling the need for developersand ops people to work together closely and cross pollinate. At the infrastructure serviceslevel itself, the role of ops got “reduced” a bit. To put it in another way, from ops being theface of IT and the go to folks for anything IT related, they were forced to take a reduced rolein the DevOps culture. But, let us keep in mind that ops are critical to the very success ofinfrastructure cloud services. The only change from the traditional era is that they have giventhe limelight to service interfaces and do their magic (as usual) in the background helpingthe cloud service providers run their infrastructure smoothly. Not only they have faded intothe background, the number of ops people needed to run the infrastructure got drasticallyreduced due to the automation at scale. Cloud services pushed Ops from being the face ofIT to the invisible face of IT. PaaS takes this one step further by making even DevOps lessrelevant because the PaaS providers absorb almost all of the operations underneath and
  14. 14. offer a simple interface for developers to deploy their apps. Hence, we are seeing the raiseof the term NoOps.Makes sense. Why are Ops people whining then?Well, the reality is not so simple and there are many shades of grey. Yes, NoOps is amarketing term but it is a great term that clearly highlights what the service is offering. Ifterms like converged infrastructure, cloud computing, DevOps, etc. can be valid terms todescribe the respective offerings, NoOps is a very legitimate term to define what hostedPaaS offers organizations. However, the ground reality is different from a simplistic evolutionto hosted PaaS. In the next several years, we are going to see a complex evolution withmost of the workloads moving to clouds while some of them staying inside the firewall. Also,we are going to see a more federated ecosystem of infrastructure players. We will be seeingadoption of both hosted PaaS as well as Private PaaS (yet another term used to describethe platform layer put on top of private cloud infrastructure). All these different choices aregoing to give us an environment where the relevance of Ops will be visible in some casesand invisible in others. In most cases, ops people will be in the background (on the serviceprovider side) and doing their magic quietly and the service interface is going to be thefuture face of IT. DevOps will stay put as long as we have organizations wanting to havemuch deeper control over the infrastructure and even in the case of hosted PaaS, somedevelopers may need to assume operational responsibilities (albeit, very rarely). Since theground reality is a bit more complex and the NoOps term completely sweeps away the realityunder the carpet, people are getting upset. But it is time for them to get used to the term andthe decibel levels are going to rise as more and more organizations start embracing PaaS.If buzz words can be a competitive advantage in a free market, NoOps is as legitimate asother terms like DevOps or Cloud. #justsayinIBM Research Shows How the Cloud isDriving Business Model InnovationCMS WireBy Barb Mosher (@barbmosher) Mar 8, 2012The decision to move to the cloud has traditionally been about operational efficiency, butaccording to IBMs research, well soon start to see organizations take advantage of thecloud for business initiatives and that kind of stuff is a whole lot more fun.We had the opportunity to talk about the IBM study with one its authors, Saul Berman,Global Lead Partner for Strategy Consulting and Innovation and Growth for IBM GlobalBusiness Services. Its important to point out that this study was conducted from abusiness perspective and not a technology perspective, which is a refreshing approach tounderstanding how the cloud can work for your business.Editors Note: Read the full study: The power of cloud. Driving business model innovation(1.35 MB PDF)The Power of the Cloud IBM StudyThis study was conducted through the IBM Institute for Business Value, in conjunction withthe Economist Intelligence Unit. It included 572 business and technology executives acrossthe world, in organizations ranging from large (greater than US$ 20 billion) to small (lessthan US$ 1 billion). The results?That, although many organizations focus cloud initiatives on operational efficiency, wellsee that slowly decrease over the next few years (from 55% to 31%) in favor of innovativebusiness plans, like new lines of business/industries, new pricing models and better partnercollaboration.
  15. 15. When surveyed, heres what items topped the cloud adoption list for most organizations:Tapping Into That PowerOne issue that has the ability to slow down the use of cloud models for innovation is thatmany organizations still see the cloud as an IT solution. But as its importance starts to reachfurther into the business, the opportunities available are being recognized.IBM notes six "game-changing" business enablers that will transform how organizationsleverage the cloud for business innovation, shown below:Cloud Enablement Framework
  16. 16. Its not a cloud maturity model, you arent going to move your organization through eachphase:IBM Cloud Enablement FrameworkThe framework looks at two things: the customer value proposition and the value chain.Along each of these dimensions there are different types of organization models. You needto look internally and decide where in this model your organization fits. Are you: ● An Optimizer: Optimizers are about enhancing what they have now and improving operational efficiency. They arent ready to take the risks and therefore wont get the revenue and market share gains that Innovators or Disruptors will. But the opportunities are there to deepen relationships with customers and enhance products and services. ● An Innovator: Innovators take advantage of the cloud to greatly extend the value proposition. This can change their role in the industry and/or lead to new markets or industries. Its about extending what they have and transforming in ways that lead to new revenue streams and market opportunities, thus gaining competitive advantage. ● A Disruptor: For a Disruptor, its about radical change, creating new markets/ industries or disrupting existing ones. Its big risk for big reward.
  17. 17. Berman says you can choose to evolve over time from one type to the next, but someorganizations are going to be innovative or disruptive from day one. He points to the media/entertainment industry as an example of an industry where you might want to focus more oninnovation and disruption than worry about an existing business model that may be underattack.Cloud Computing Goes Far BeyondVirtualizationVirtualization vs. Private Cloud (Part 1)Virtualization JournalBy Yung ChouVirtualization vs. private cloud has confused many IT pros. Are they the same? Or different?In what way and how? We have already virtualized most of my computing resources, is aprivate cloud still relevant to us? These are questions I have been frequently asked. Beforegetting the answers, in the first article of this two-part series listed below lets first go througha few concepts.Part 1: Cloud Computing Goes Far Beyond Virtualization (This article)Part 2: A Private Cloud Delivers IT as a ServiceLately, many IT shops have introduced virtualization intoexisting computing environmentw. Consolidatingservers, mimicking production environment, virtualizingtest networks, securing resources with honey pots,adding disaster recovery options, etc. are just a fewapplications of employing virtualization. Some also runhighly virtualized IT with automation provided by systemmanagement solutions. I imagine many IT prosrecognize the benefits of virtualization including betterutilization of servers, associated savings by reducingthe physical footprint, etc. Now we are moving into a cloud era, the question then becomes "Is virtualization the same with a private cloud?" or "We are already running a highlyvirtualized computing today, do we still need a private cloud?"The answers to thesequestions should always start with "What business problems are you trying to address?"Then assess if a private cloud solution can fundamentally solve the problem, or perhapsvirtualization is sufficient. This is of course assuming there is a clear understanding of whatis virtualization and what is a private cloud. This point is that virtualization and cloudcomputing are not the same. They address IT challenges in different dimensions andoperated in different scopes with different levels of impact on a business.VirtualizationTo make a long story short, virtualizationin the context of IT is to "isolate" computingresources such that an object (i.e. an application, a task, a component) in a layer abovecan be possibly operated without a concern of those changes made in the layers below.A lengthy discussion of virtualization is beyond the scope of this article. Nonetheless, letme point out that the terms, virtualization, and "isolation" are chosen for specific reasonssince there are technical discrepancies between "virtualization" and "emulation", "isolation"and "redirection." Virtualization isolates computing resources, hence offers an opportunity torelocate and consolidate isolated resources for better utilization and higher efficiency.
  18. 18. Cloud ComputingCloud computing on the other hand is an ability to make resources available on demand.There are statements made on what to expect in general from cloud computing. A definitionof cloud computing published in NIST SP-800-145 outlines the essential characteristics, howto deliver, and what kind of deployment models to be cloud-qualified. Chou further simplifiesit and offers a plain and simple way to describe cloud computing with the 5-3-2 Principle asillustrated below.The essence of cloud computing is rooted at the appreciation of a "service." In the context ofcloud computing, a service simply means the state of being available on demand.So SaaS means software, i.e. an application, is available on demand and the focus ison functions available within and not beyond the application. PaaS provides a run-timeenvironment on demand and the scope becomes what are the common set of capabilitiesavailable on demand for applications deployed to this run-time environment. Since therun-time environment is available on demand, an application deployed to the run-timeenvironment then can be brought to a running state on demand. Namely those applicationsdeployed to a PaaS environment are delivered, as a consequence, with SaaS. AndIaaS denotes infrastructure available on demand, which means the ability to provisioninfrastructure on demand. For IT professionals, provisioning infrastructure at an operationallevel translates to deploying servers. And in the context of cloud computing, all servers arevirtualized and deployed in the form of a virtual machines or VMs. So, IaaS ultimately is theability to deploy VMs on demand. "On-demand" is not to be casually used. This is a loaded term with a strong connotation of the five essential characteristics of cloud computing. On-demand means high accessibility and always-on readiness since it must be accessible and ready per SLA. In cloud, they are represented by self-service model and
  19. 19. ubiquitous access. On-demand suggests there are likely standardization, automation, optimization, and orchestration in place, which are presented collectively as resource pooling and elasticity. On-demand implies the need for auditing and metering, i.e. analytics, so capacity can be planned accordingly. And that is why consumption-based charge-back or show-back model is included in the essential characteristics of cloud computing.Unequivocally DifferentWith what has been described above, to realize the fundamental differences betweenvirtualization and private cloud becomes rather straightforward. Noticeably, virtualization isnot based on the 5-3-2 principle as opposed to cloud computing is. For instance, a self-serving model is not an essential component in virtualization, while it is essential in cloudcomputing. One can certainly argue some virtualization solution may include a self-servingcomponent. The point is that self-service is not a necessary , nor sufficient condition forvirtualization. While in cloud computing, self-service is a crucial concept to deliver anytimeavailability to user, which is what a service is all about. Furthermore, self-service is an effective mechanism to in the long run reduce training and support at all levels. It is a crucial vehicle to accelerate the ROI of a cloud computing solution and make it sustainable in the long run. Virtualization is centered on virtual machines and rooted in infrastructure management, operations, and deployment flexibility. Virtualizationis about the abilities to consolidating servers, managing VMs, streaming desktops, and so on. How to productively configure, deploy, and manage aworkload in a deployed VM andAt the same time, cloud is about "service"and "service" is about the readiness andresponsiveness relevant to market opportunities. Cloud is about go-to-market. Cloud focuseson making a requested LOB application available on demand and not just on just how todeploy a VM. Cloud is interested in not only operating VMs, but providing insights of a targetapplication running in those VMs.No, virtualization is not cloud computing. And cloud goes far beyond virtualization. So whatare the specifics about virtualization vs. a private cloud? [To be continued in part 2]Yung Chou is currently a Sr. IT Pro Evangelist in Microsoft. Within the company, he has hadopportunities serving customers in the areas of support account management, technicalsupport, technical sales, and evangelism. Prior to Microsoft, he had established capacitiesin system programming, application development, consulting services, and IT management.His recent technical focuses have been in virtualization and cloud computing with stronginterests in private cloud with service-based deployment and emerging enterprise computingarchitecture. He is a frequent speaker in Microsoft conferences, roadshow, and TechNetevents.Top 5 Things The Cloud Is NotPeder Ulanderposted in Blog, Featured ⋅ June 22, 2012 1:57 pm
  20. 20. The cloud looks set to be the next king, but there are five things the cloud is not. Can youthink of others? Image: akakumo/FlickrIt’s clear that the technology industry is moving from the PC era to the cloud era in severalsignificant ways. While cloud represents a new way for IT to deliver — and end users toconsume — IT applications and services, this transition also represents a significant changein how applications, services and systems are defined. The move to cloud computing is themost important technology disruption since the transition from mainframe to client-server, oreven since Al Gore invented the internet. While industry veterans like Oracle’s commanderin chief declared it a fad, this is a decade-long trend that is here to stay, and one that willdefine the next generation of IT.The movement itself has been in play for the last decade, however there continues to be alot of (mis)information in the marketplace about the cloud. So much so that it is difficult fororganizations to figure out what is real and what is not to help them develop a successfulcloud strategy, or simply learn about technologies that have been specifically designed andpurpose-built to meet this dramatic shift in technology. While it’s important to know what thecloud is, it’s just as important to separate the wheat from the chaff, and for IT to understandwhat cloud is not.To this end, I encourage you not to add yet another definition of the cloud to your glossary,but to truly understand the top 5 things the cloud is not.1. Cloud is not a place.People often talk about moving to the cloud as if they were moving to another city. But thecloud is not a place. In fact, the cloud can be anywhere, in your data center or someoneelse’s. Organizations that believe they are moving to a strategy that leaves legacy apps andsystems behind are in for a rude awakening. The single most important way for enterpriseorganizations to prepare themselves for the cloud is to understand that the cloud is aradically new way of delivering, consuming and adopting IT services in a far more agile,efficient, and cost-effective manner, which will spread throughout the ether and be a mix ofpublic, private, managed or hybrid clouds. By looking holistically at the cloud, organizationscan optimize its benefits for their budgets, privacy needs, geographies and overall businessneeds.
  21. 21. 2. Cloud is not server virtualization.Despite what many believe, and what many will tell you, the cloud is not the same as next-gen server virtualization. It doesn’t surprise me that many believe that by virtualizing theirdata center they will create a private cloud. Some vendors are intentionally trying to blur thatline, aiming to convince customers that their vCenter clusters somehow deliver a privatecloud. On the contrary, that is a gross exaggeration of the term cloud.If you take a look at the way Amazon has built its cloud architecture, it becomes very clearthat there are some fairly stark differences between a server virtualization environment anda true cloud architecture. While Amazon starts with Xen virtualization technology, the brainsof its architecture comes with a new layer of software that Amazon built in an effort to createa new control plane, a new cloud orchestration layer that can manage all the infrastructureresources (compute, storage, networking) across all of their data centers. This is at theheart of the cloud’s technology disruption. Some analysts refer to this as the “hypervisor ofhypervisors,” or a “new software category of cloud system software.”The fact of the matter is that some of the major players are doing cloud without servervirtualization. Take Google for example. They have deployed a cloud architecture that is notusing server virtualization, but rather a bare metal infrastructure. So while virtualization canbe an important ingredient of cloud, it is not always a requirement.3. Cloud is not an island.Depending on what you’re reading, you’ll hear a lot about public clouds versus privateclouds, and it may feel as if enterprises must make a wholesale decision on which way togo. But the cloud is not an island, it is not a place where you put all of your IT services, andthen lose all interconnectivity and access. The recent Amazon outages have proven this tobe an important point for any organization leveraging the cloud. The right cloud strategy willbe one that enables you to have a hybrid approach with the ability to easily connect privateand public clouds. Even the recent move by NASA to include Amazon Web Services as partof its cloud rollout after a significant investment in the build-out of its own technology provesthat the market is moving to open, interoperable multi-cloud environments.4. Cloud is not top-down.The cloud has up-ended the traditional IT approach to delivering services. The lines ofbusiness have been leading the charge in making the decision to move to cloud computing.With specific needs to get to market quickly, functional business leaders are consumingcloud services to avoid traditional IT processes. But we don’t need surveys to clarify thismovement. The reality is that with the simple swipe of a credit card and the creation of anaccount, end users can gain instant access to infinite pools of IT resource to help test out anew idea, get their job done or even become more agile in their daily work. This is part ofwhy this revolution is so powerful. The Consumerization of IT is driving this new movement.Users are already there and the C-level offices are just now trying to catch up with them.Those that embrace this move sooner rather than later will learn how to use the cloud asa strategic weapon before their competitors do. So the cloud is not top down, but rather abottoms-up phenomenon.5. Cloud is not hype.As I started this piece, I wrote about the (mis)information that has flooded the marketand slowed progression and adoption of the cloud for some organizations. I’ve spokenwith people in many organizations who are still skeptical of the cloud and believe that it issomething that is very far off into the future. No doubt there is a lot of noise in the marketwith many claiming early victory in the hearts and minds of developers, with open sourcemomentum, or beta products. The reality is that the cloud is ready now, and Citrix hasmore than 100 organizations that are running clouds in production today. Companieslike AutoDesk, Edmunds.com, Nokia, Chatham Financial and others, already reaping thebenefits.
  22. 22. My words of advice to companies considering a move to the cloud – learn from others whohave already built highly scalable, successful clouds that have helped them transform theway they deliver and consume IT resources.This is just the beginning of the discussion. There are many more topics that we will continueto talk about in the coming weeks, months, years (such as, cloud is not only an infrastructureand cloud is not just for service providers). All with the goal of helping organizations and themarket understand what the cloud is and what it is not.Peder Ulander is vice president of product marketing for the Cloud Platforms Group atCitrix, overseeing the company’s marketing strategy for its cloud infrastructure and servervirtualization products.CIOs Dont Need to Be BusinessLeadersGiven the complexity of todays applications, itsfolly to suggest that the future role of the CIO is lesstechnical and more businesslike, columnist BernardGolden writes. If anything, its the opposite -- thebusiness side of the enterprise should embracetechnology.By Bernard GoldenFri, May 18, 2012It seems like every week I come across an article stating that being a CIO means thinkingmore like a business person and less like an engineer. Often I see articles that say that CIOsneed to talk the language of business, not technology. Occasionally Ill see one that saysthat CIOs need to be business leaders and stop focusing on technology.I have seen pieces asserting that future heads of IT will be from disciplines such asmarketing or finance, since technology really isnt that important anymore. Ive even seenanalyses that say that CIOs no longer need to manage technically capable organizationsbecause infrastructure is being offloaded to outsourcers and on-premise applications arebeing displaced by SaaS applications.The implication of all these viewpoints is that technology qua technology is no longersignificant and that, overall, its so standardized and commoditized that it can be treated likeany other area of the business. In fact, it can be managed by someone with no technicalbackground at all.The general rap against technical IT executives is that they talk about technology too muchand fail to communicate with CEOs in so-called "business terms." The thinking is that CIOsfail to use the language of business and thereby bore—or, worse, alienate—CEOs, with theresult that CIOs are dismissed from the inner ranks of corporations.If only CIOs could learn to communicate in business terms, the argument goes, then theywould be accepted into the inner circle, embraced by CEOs no longer discomfited bytechnical jargon.
  23. 23. Notion of CIO as Business Leader Just Plain WrongThe shorthand version of this argument is the CIO needs to be a business leader, not atechnologist. The implication is clear: The CIO leaves the technical details to others andfocuses on the big picture.Theres only one thing wrong with this perspective. Its wrong. In fact, nothing could befurther from the truth.Technical skills in IT management are important today like never before—and that fact isbecoming increasingly evident. In the future, CIOs will need deep technical skills. A CIOwith even average technical skills will be not only inadequate for his or her job, he or she willrepresent a danger to the overall health of the company.Frankly, even on its surface, this argument of "CIO as business leader" doesnt make sense.Marketing, for example, is undergoing radical transformation as it shifts to online and digital.Today sophisticated analysis of click patterns, A/B testing, big data analytics and so onare a core marketing competence. Do you think that CEOs want a head of marketing whodoesnt know the details of how these kind of marketing tools operate? That marketing is runby someone who can use the language of business, even though he or she doesnt reallyunderstand the details of what is done in the marketing programs? Of course not.IT, too, is becoming increasingly complex. Ten years ago, a companys website wasprimarily a display application designed to deliver static content. Today, a website is atransaction and collaboration application that supports far higher workloads. Websitescommonly integrate external services that deliver content or data that is mixed with acompanys own data to present a customized view to individual users. The application mayexpose APIs to allow other organizations to integrate it with their applications, and thosesame APIs may be used to support a mobile website. Finally, the site probably experienceshigh variability of load throughout the year as seasonal events or specific business initiativesdrive large volumes of traffic.Application performance management depends on an exquisite tuning of a multitude ofelements, any of which can affect response time and each of which must be monitored toassess an apps ongoing health. To be sure, one can expect the application to constantlychange as new business arrangements, partnerships, or corporate events such as mergersor acquisitions require functionality changes.The complexities of these applications is of an order of magnitude higher than those ofa decade ago. For a discussion of what these applications look like from an enterprisearchitecture perspective, read this post by friend and colleague James Urquhart and just tryto come away thinking that this highly complex, dynamic, constantly evolving environmentcan be managed by someone without technology chops.You Cant Discuss Tech Without Knowing TechHeres the thing: Complex as they are, these new applications are critical to the success ofthe overall business. The website of 2000 was important, but if it wasnt operating properly,the company could still function. If todays Web-enabled application isnt available, businessgrinds to a halt. This reflects how, over time, these applications have insinuated themselvesinto the core functionality of the company—and made their successful operation critical tothe operation of the business.Now, do you think a CIO can get by without understanding the key elements of these typeof applications? Without recognizing the weak aspects of the application where failure orperformance bottlenecks can ruin successful user engagement with the application?The counter argument to this perspective is that the technology is too complex andBelieveme, there is a world of difference between someone who understands technology—andas a result has to weigh alternatives and disputes among different groups involved in atechnology discussion—and someone who doesnt really have any technology background
  24. 24. and arbitrates by non-technical criteria. The difference between them is the differencebetween an organization that gets things right on technology—or, when it gets things wrong,can recognize the issue and quickly correct it—and one that makes poor decisions thatresult in fragile, constrained applications.In Todays Economy, CEOs Obligated to Know TechFrankly, that issue of talking to the CEO in business language with which he or she iscomfortable is a red herring. The fact is, businesses today are technology businesses.Information technology is core to what they do. Something so critical to a companyssuccess imposes an obligation on a CEO to comprehend it. After all, do you think theCEO of GM refuses to engage with the head of manufacturing on supply chain issueseven though its a highly technical subject? Why, then, is it OK for a CEO to deflect an ITdiscussion because its highly technical?Now that I think about it, it might be time to turn the whole argument on its head. Thestatement shouldnt be that CIOs arent businesslike enough. Its that too many of todaysCEOs are insufficiently technical.Bernard Golden is CEO of consulting firm HyperStratus, which specializes in virtualization,cloud computing and related issues. He is also the author of "Virtualization for Dummies,"the best-selling book on virtualization to date. Most recently Wired.com namd him one ofthe Top 10 Cloud Influencers and Thought Leaders. Follow Bernard Golden on Twitter@bernardgolden.Follow everything from CIO.com on Twitter @CIOonline, on Facebook, and on Google +.