Network Computing is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

10 Tools To Prevent Cloud Vendor Lock-in


  • In enterprise computing, vendor lock-in is too often a fait accompli. Vendor lock-in happens when, for example, a particular company -- such as IBM, Microsoft or Cisco Systems -- becomes the dominant vendor behind a particular technology and develops products that capture the advance with proprietary elements. That prevents its customers from leaving and ensures that only proprietary vendors can continue to capitalize on the technology.

    With IBM it was the mainframe. Microsoft for many years dominated end-user computing with its Windows operating system, and Cisco has become the dominant enterprise networking vendor in the Ethernet era.

    With the advent of cloud computing, however, customers can avoid lock-in by taking back some of that decision-making power. The cloud is a highly standardized environment, with new standards being added all the time -- such as the DMTF's Open Virtualization Format 2.0 (OVF) -- allowing translation between the proprietary virtual machine formats.

    One way to get locked into the cloud is to, say, use only tools that recognize VMware virtual machines or Amazon Web Services VMs based on Amazon Machine Images. These virtual file formats are proprietary. They can be converted and moved around, but you need the right tools to do so. Faced with the prospect that customers might do it themselves, Amazon Web Services has moved beyond recognizing only its own virtual machine format to recognize VMware's as well. VMware and Microsoft recognize and produce some tools that work with each other's file formats. The old lock-in bonds are already being eroded by the nature of cloud computing, and this process will continue.

    While there are still many proprietary traps in the cloud -- for example, using only Microsoft .net languages and running workloads only on the Windows Azure cloud -- you don't have to be hemmed in. Tools that recognize .net languages but deploy to public clouds are available. Don't let the convenience of vendor familiarity lead you to lock-in. Cloud computing isn't owned by anyone -- on the contrary, it can and will be the foundation on which many vendors will compete.

    Furthermore, enterprise cloud users are eager to tap into a variety of services. They're building private clouds on premises to hold their most critical data, but they also want to interoperate with public clouds for customer-facing activities. That combination is called hybrid cloud computing, and there are multiple tools available.

    In general, the best tools go beyond the operating infrastructure to view how well separate servers are running. They provide a view of the application, which uses pooled, or virtualized, resources such as storage and networking. This application-centric point of view addresses such questions as: How well is the application running? Do you need to perform analytics to see if something is going wrong? Do you need to scale it out further? How are the end users seeing the application?

    Keeping choice in the hands of the IT manager is an important part of cloud computing. Here are 10 tools that can help you avoid lock-in and keep your enterprise moving forward.


  • Enstratus provides a management console that works across more public cloud services than many of its competitors. It configures your workload in a neutral software layer using open source Puppet or Chef, then translates beneath the covers to move into a public cloud. It can move data to Amazon Web Services, Bluelock, GoGrid, CloudSigma, Eucalyptus Community Cloud, HP Cloud, Rackspace and others.

    The Enstratus console gives its operators a view of both internal private cloud and external public cloud workloads. The console will also connect to an additional list of cloud storage services, such as Google Storage for Developers or EMC Atmos. Its latest supported public cloud infrastructure is IBM's SmartCloud Enterprise.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • BMC is a systems management software company with roots in the mainframe world. Offering troubleshooting diagnostic capabilities, BMC has quietly realigned itself as a vendor-neutral management system for cloud computing. It started out with a market-leading configuration management database product, Atrium, and built from there on top of its BladeLogic x86 cluster automation system. Its Cloud Lifecycle Management product can provision virtual machines from any of the major virtualization vendors and assign them storage and networking, allow for end-user self-service from a catalog, manage by policies, create blueprints of reusable software stacks, implement user chargeback and monitor the resulting environment.

    Its sister product, BMC Cloud Operations Management, adds ongoing operational know-how. It can look across a variety of hardware and virtualized resources to spot slowdowns and trouble spots, provide root-cause analysis, and use predictive analytics to spot the next hotspot or perform capacity management. Its latest version, COM 9.0, pushes its cloud management into hybrid cloud operations, with the data center extended into the public cloud. Unlike some tools, it can focus on application performance as well, with key information derived from Atrium, and provide policy-driven placement of workloads on cloud servers.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • Abiquo, founded in 2006 in Redwood City, Calif., caught the emergence of the cloud wave at the right time, with its emphasis on private/public cloud interoperability and mobility between public cloud services. Abiquo Enterprise Edition is a Java system that can generate and deploy workloads for multiple cloud environments. It works with virtual machines generated by major hypervisors: VMware's ESX Server, Citrix Systems XenServer, Microsoft's Hyper-V, open source KVM and Xen, and Oracle's OracleVM. A workload deployment and management system for private clouds powered by those hypervisors, it's also capable of deploying to the public cloud and allows IT managers to view all workloads on a single pane of glass.

    It includes open source JasperSoft reporting and OpsCode Chef recipes for end-user self-provisioning. Specifically geared to work with Cisco Systems' Unified Computing System servers often deployed in heavily virtualized environments, it can assign a service profile template to each UCS blade in a rack, providing the computing characteristics needed by the workloads it will serve.

    Public clouds with which it connects include mainstream Amazon Web Services, Google App Engine, Microsoft Windows Azure, Engine Yard, Salesforce.com's Force.com, Heroku and some lesser-known clouds, such as GreenQloud. For international operations, the Abiquo user interface supports eight languages.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • ServiceMesh attempts to incorporate the frequent updates of the DevOps process into the management of cloud applications and cloud services. It does this through its ServiceMesh Agility Platform for both developers and cloud operations managers. ServiceMesh offers strong front-end features, including a Planner component, which evaluates what type of workload an application represents; a Designer component that is a graphical software stack assembly bench to produce a workload, template or application made up of several different languages; an end-user self-provisioning component called the ServiceMesh Store; and a Release Manager, a dashboard illustrating deployment environments throughout the application's lifecycle. ServiceMesh weighs in on the ops side with an Operations component for monitoring and reporting.

    Regardless of where an application is to be deployed -- private cloud, public cloud or a combination -- that deployment can be automated on the platform. Identical security and compliance policies follow the application, regardless of where it runs. As applications are updated they are automatically released into cloud environments, and enforcement of service agreements can be provided. ServiceMesh can be used to manage Vblocks, the virtualized racks of servers from the VMware, EMC and Cisco subsidiary Virtual Computing Environments.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • RightScale was an early online service for provisioning, deploying and managing workloads across a variety of public and private clouds. Since 2007, it's concentrated on hiding as much of the configuration and deployment process as possible in its RightScale Cloud Management systems so customers don't see or worry about it -- it handles issues related to virtual machine formats or conversion processes.

    Co-founder and CEO Michael Crandell has become an advocate of customers using multiple public clouds and selecting the best target environment for their workloads based on the platform's ability to manage the migrations. RightScale deploys workloads to SoftLayer, Rackspace, AWS, Windows Azure, Datapipe and HP Cloud, among others. For customers looking to manage on-premises private clouds, RightScale can deploy to the big three open source internal clouds: Eucalyptus, CloudStack and OpenStack. Once a workload is deployed, RightScale will monitor operations, and its automated load balancing will handle scaling needs according to the customer's parameters. Customers get their own dashboard views of running systems and can modify their operational guidelines as they see fit.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • Scalr is a workload management system that started as an open source project by MediaPlug, a media-sharing service, for sending workloads to Amazon Web Services. It's now an Apache 2 open source code project at Google Code, with an enterprise version offered as a hosted service by startup Scalr.net, founded by Sebastian Stadil. It can provision servers in EC2, scale out servers as traffic demands, perform disaster recovery and allow changes in server management. Although Scalr has been described as "a poor man's RightScale," its capabilities have gotten increasingly sophisticated. It can recover a stalled server, and scale and recover database servers as well as regular application workloads.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • Gravitant is an example of what may become a new way of distributing cloud services: a cloud service brokerage. Gravitant's Cloud Matrix CSB Platform is a brokerage and management platform for cloud services, allowing a company to contract with a variety of public and private service providers integrated into one cataloging, provisioning and billing portal. It is service-agnostic, and a company can simulate various mixes of cloud services to explore different pricing scenarios. With CloudMatrix CSB, a company can set up a virtual data center that uses several cloud services while managing the data center as a single unit. Gravitant is one of the first companies to offer such a brokerage service.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • Stamford, Conn.-based Kaavo takes a unique approach: Instead of focusing on cloud infrastructure, it focuses on application lifecycle management – regardless of whether the application is found on a public, private or hybrid cloud. To do this, it uses its own system definition file, a single XML file that captures a description of both the components and orchestration sequence of a complex application.

    First included in 2009 in Kaavo's flagship IMOD workload deployment and orchestration engine, the use of the definition file allows IMOD to automatically prepare the workload for deployment to any of the major cloud environments. IMOD can simplify application code changes, updates and deployments down to a few clicks in the management console. It also tracks all changes to the definition file.

    The definition file can invoke a workflow engine to perform scheduled events such as deployments. It can perform automated scaling and automated setup of secure network connections as needed.

    Kaavo is intended to manage the lifecycle of the application as a cloud workload, so it can be used to govern development and testing as well as production deployment. It tracks hours of use for billing and chargeback. It can deploy to multiple cloud environments, but only one view of the application is maintained. It works with HP Cloud Services, Amazon Web Services EC2, IBM Enterprise SmartCloud, Logicworks, Rackspace, Terremark and VMware's vCloud Director-equipped clouds.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths


  • VMTurbo's Operations Manager allows a multi-vendor, on-premises private cloud to manage virtual machines from different sources. It supports VMware's vSphere and VCloud Suite, Microsoft's Windows Server 2012 with Hyper-V 2012, Red Hat Enterprise Virtualization with KVM and Citrix XenServer. It also supports CloudStack, an early open source code implementation of a cloud software stack based on OpenStack.

    The Burlington, Mass.-based company has built an applications-centric point of view into Operations Manager, while allowing the application to run in a variety of virtual machine formats. VMTurbo collects data from the running virtual machines to monitor application performance and health and can use analytics to determine whether an application has the server capacity it needs.

    The analytics engine can apply business rules and policies to show a dashboard on a server cluster that sums up existing workload needs, remaining capacity and projected future capacity needed. VMTurbo offers a cross-hypervisor view while maintaining demand trends and capacity planning tools.

    Recommended Reading

    Cloud Implementation Costs, Complexity Surprise Companies

    Hybrid Cloud Management Is Within Reach

    8 Cloud Tools For Road Warriors

    Are Big Companies Ready For Cloud ERP?

    OVF 2.0 Takes On Cloud Lock-In Worries

    7 Dumb Cloud Computing Myths