Microsoft recently announced support for Docker and Kubernetes technology on the Azure platform. Docker is an open source technology that enables developers to “build, ship, and run distributed applications” by means of container technology that facilitates application migration and deployment. Meanwhile, Kubernetes is an open source cluster management platform that can be used to deploy Docker containers. Azure customers can now use the Kubernetes platform to create and publish Docker containers to the Azure storage platform. In addition, Azure customers can deploy and configure Azure clusters using container images from Azure Storage or the Docker Hub. The Microsoft Azure team also built the Azure Kubernetes Visualizer which provides developers with a visual representation of the status of Kubernetes when managing Docker technology as illustrated below:
The Azure Kubernetes Visualizer is intended to “visually demonstrate some Docker and Kubernetes concepts such as containers, pods, labels, minions, and replication controllers.” Given that Kubernetes was open-sourced by Google in June, Microsoft’s decision to support Kubernetes and Docker represents a stunning example of the way in which container technology promises to drive product development as it relates to Infrastructure as a Service cloud computing. Amazon Web Services, for example, has yet to embrace Docker technology, and given its history of ignoring open source projects with the exception of Apache Hadoop, is likely to continue to withholding support for Docker. Meanwhile, Microsoft’s decision to support Kubernetes in conjunction with Google and its IaaS Google Compute Engine opens the door for the possibility of increased inter-operability between Azure and GCE, and subsequently promises to assert the criticality of Docker and Kubernetes to conversations about cloud computing inter-operability. The real winner from Azure’s support for Kubernetes and Docker, however, is Docker, whose container technology is likely to continue skyrocketing in adoption and affecting market dynamics not only within the IaaS space, but also the Platform as a Service (PaaS) landscape given its ability to reduce the gap between application development and operations by streamlining application migration across different cloud environments.
On Thursday, Amazon Web Services (AWS) announced it had become the first cloud vendor to obtain provisional authorization from the Department information Systems Agency (DISA) to handle DISA data requiring levels 3-5 security clearance under the DISA Cloud Security Model (DISA CSM). Because AWS had already achieved authorization to manage data for levels 1-2 security clearance as of March, Thursday’s announcement means that AWS can now handle all unclassified data given that level 6 security clearance applies to classified data. The DISA CSM imposes even more restrictions on cloud-based data than the FedRAMP certification program. Department of Defense Agencies can now leverage the capabilities of the AWS GovCloud more quickly given the expanded scope of its authorization to handle a greater range of unclassified data. As the very first cloud vendor to obtain DISA clearance for levels 1-5 of unclassified data, AWS stands poised to strengthen its market share advantage in the battle for government-based cloud services given its receipt of a hotly contested contract for $600M and leadership with respect to obtaining FedRAMP certification.
Rackspace recently announced details of a managed cloud service plan that gives customers the opportunity to take advantage of managed services for their cloud deployments. The managed cloud service plan comes in two forms: (1) managed infrastructure, which provides advisory services regarding infrastructure set-up and architecture; and (2) managed operations, which enables Rackspace engineers to access customer servers to tweak code as necessary. The managed infrastructure and operations offerings represent Rackspace’s attempt to differentiate itself from competitors such as Amazon Web Services and Windows Azure, both of which demand greater responsibility on the part of developers and IT staff to provision, configure, deploy and manage Infrastructure as a Service environments. The introduction of the managed cloud service pivots on Rackspace’s famed “fanatical support” by building on the company’s strengths as a leader in consultative support for IaaS deployment and management. Rackspace President Taylor Rhodes summarized the new managed cloud offerings as follows:
Our basic level, called Managed Infrastructure, offers Fanatical Support with much more managed service than do the more-expensive, premium service levels offered by many of our competitors. Our higher service level, called Managed Operations, provides even more managed services, up the stack into the support of application level — addressing customer needs that most of our rivals won’t even touch.
Components of the managed infrastructure offering include architectural advice, support for workload migration and scaling, launch assistance and round the clock availability of cloud engineers to troubleshoot and resolve issues. Managed operations additionally delivers support for operating systems, web servers, database servers, cloud databases, cloud backup and monitoring and user provisioning and permissions. Rackspace’s managed infrastructure offering is priced at $.005/GB, assuming a $50 minimum per month while managed operations is priced at $.02/GB, with a $500 monthly minimum. In addition to its managed cloud service, Rackspace announced details of an expanded program for developers and more transparent pricing. Altogether, Rackspace’s new managed cloud offering is likely to give it some short term publicity and inject new life into its ailing IaaS positioning, but the San Antonio-based company will need a deeper transformation if it intends to seriously compete with the big players in the IaaS space, particularly given that competitors such as Amazon Web Services already partner with other vendors to offer managed services comparable to those revealed by Rackspace last Tuesday.
On Tuesday, Virtustream announced the release of xStream 3.0, the cloud management platform that allows cloud service providers to create public clouds and enterprises to create private clouds. Based on Virtustream’s proprietary µVM™ technology, the xStream platform empowers customers to create secure, IaaS environments for mission-critical applications that can tolerate little or no system downtime. Virtustream’s xStream technology is engineered to minimize costs associated with migrating applications to the cloud and streamline the day to day management of the subsequent cloud deployment. Available as both software and an appliance, the xStream platform has evolved in its latest release to target government, cloud service provider and enterprise customers with stringent demands for availability and compliance. xStream 3.0 features enhanced enterprise risk management (ERM) functionality that allows customers to implement ERM solutions not only for an organization’s IT infrastructure, but also for the organization as a whole. In addition, xStream 3.0 delivers support for OpenStack, enhanced data protection functionality, application performance management monitoring tools and predictive analytics that optimize infrastructure management. With this release, Virtustream announces itself as a major player in the market for production-grade cloud deployments that require rich and nuanced ERM and data security functionality. As ERM implementation proliferates in verticals outside of financial services and the government sector, expect cloud technologies such as xStream to enjoy increasing adoption as a result of its suite of tools for compliance, data protection, application monitoring and infrastructure optimization. xStream’s support of both OpenStack and VMware represents yet another feather in its cap whose importance will reveal itself as OpenStack technology continues to assert its imprint on the IaaS space in conjunction with OpenStack’s maturation.
The screenshot below of the xStream management console illustrates the platform’s ability to track resource consumption, storage and costs:
On Monday, IBM opened a new SoftLayer datacenter in Hong Kong as part of a $1.2 billion investment to strengthen its cloud services in Asia and all over the world. The Hong datacenter represents the first of 15 datacenters that IBM plans to open worldwide this year, bringing its total fleet of centers to 40 by the end of the year. Separate from the Hong Kong datacenter, IBM currently has 13 SoftLayer data centers and 12 from IBM. The Hong Kong data center has capacity for over 15,000 servers and complements IBM’s presence in Asia by way of the Singapore datacenter and “network points of presence” in Hong Kong, Singapore and Tokyo. Moreover, it positions SoftLayer to serve the entrepreneurial community’s cloud services needs in Hong Kong. IBM intends to deploy data centers in all major geographic regions and financial centers worldwide, including plans to extend its cloud services presence to Africa and the Middle East in 2015.
Tags: Iaas, SoftLayer
This week, polyglot private PaaS vendor ActiveState announced the release of ActiveState Stackato 3.2, which features an array of enhancements that collectively focus on improving the security, scalability and ease of management of the platform. Version 3.2 delivers single sign-on functionality to ensure that access to the application complies with the identity management protocols of customers. In addition, this release upgrades the granularity of application-related permissions in order to give system administrators expanded control over who has access to specific components of the Stackato PaaS in conjunction with the ability to tweak their rights and permissions. Version 3.2 also offers placement zones and availability zones that allow customers to deploy their applications across multiple data centers or specific spheres of hardware in order to ensure the availability of applications in the event of infrastructure failover specific to one data center or set of infrastructure. Moreover, Stackato 3.2 supports application auto-scaling to maximize application availability during peak usage periods while concurrently scaling down the application during periods of diminished usage. Stackato’s auto-scaling functionality extends to Stackato on CloudStack or the Citrix CloudPlatform. Version 3.2 also features enhancements to Stackato’s management console in the form of additional dashboards that provide administrators with visual representations of the status of usage across placement and availability zones as well as memory availability and allocations across a cluster.
Stackato version 3.2 goes a long way toward rendering the Stackato platform fit for production usage amongst enterprise customers that demand granular permissions, single sign-on functionality, auto-scaling and the ability to deploy applications in multiple placement and availability zones. As such, the release constitutes yet another example of an enterprise-grade PaaS at a historical moment when PaaS technologies have been overshadowed by price wars in the IaaS space and industry innovations specific to Big Data analytics and management. Stackato, recall, is built on technology from the Cloud Foundry initiative and supports development languages such as, but not limited to, Java, Perl, Python, PHP, Ruby, Node.js, Erlang, Scala, Clojure and Mono. This week’s release illustrates the depth of innovation originating from one of the industry’s key PaaS players at a time when IaaS continues to overshadow PaaS despite the uniqueness of PaaS technology and its clear differentiation from IaaS.