In this series of articles I’ll articulate the challenges customers face in hybrid cloud adoption, the key hybrid cloud requirements and ways to address them.
Organizations are trying to transform their business and innovate faster by getting access to resources on-demand per business needs but enterprise IT has not been able to provide that. This has led to a new challenge of “shadow IT” -- employees going direct to the public cloud to get fast and easy access to resources by going around IT. Shadow IT proves that business users are looking for the flexibility of cloud, but IT is wary of public cloud due to associated concerns of security, loss of visibility and control.
Hybrid cloud enables organizations to innovate faster by enabling rapid, self-service provisioning of resources, with the choice to deploy workloads in enterprise’s own data center or in the public cloud in a pay-as-you-go and scaled out manner. Hybrid clouds enable multiple use cases such as dev/test, capacity augmentation and disaster recovery besides control of Shadow IT. There is an increased trend towards hybrid cloud as it offers flexibility to respond quickly to business needs and allows reduction in cost. As per Forrester, more then 70% of enterprises plan to complement their in-house server and storage resources with IaaS resources from public cloud providers for primary or peak workloads. This points to the fact that customers want Hybrid Clouds, not just private or IaaS public clouds.
While we see the advantages of hybrid cloud, we don’t see large-scale customer adoption yet. The factors that have been preventing this are
- No easy way to deploy and manage on-premise and public cloud resources through a single-console.
- Lack of security for workloads running in public cloud and unsecure connectivity from private to public cloud.
- Slow and complex management processes such as need to re-architect the application while migrating workloads across hybrid cloud.
Customers are also concerned about getting locked-in to a particular vendor’s solution or a particular public cloud. Today one particular public cloud may be right for certain class of applications but tomorrow another public cloud provider may offer better SLAs, cost or application performance. There are some fragmented solutions that allow migrating workloads from customer’s private cloud to a public cloud but then there is no easy way for the workloads to move back to enterprise or migrate easily to another public cloud. Customers have seen that with such solutions neither they get the complete agility nor the long term cost benefits. As a result, they are weary of getting locked-in to a particular public cloud or a solution that works only across a single hypervisor or over a certain compute, network or storage device.
Customers are looking at true hybrid cloud capabilities that means more than just running some applications on-premises and some in the public cloud. “Hybrid” clouds require a functional extension of local resources to the cloud to the same degree that local resources are connected/integrated. Lets look at the key customer requirements of a true hybrid cloud:
Self-service access: A true hybrid solution will provide self-service portal to users and IT admins. It will enable users to seamlessly deploy applications both on-premise or in the public cloud from a unified console. It will allow IT admins to manage workloads from a single pane of glass and enforce complete security for workloads in multi-tenant shared public cloud as private cloud.
Workload portability: It will offer bi-directional migration of workloads from private to public cloud independent of the underlying architecture.
IT as a broker: It will enable IT to act as a broker on behalf of Line of business while minimizing risk by enforcing that application network and security policies are identical regardless of the location of the workloads.
Open architecture: It will enable choice and flexibility for users, IT admins and cloud service providers by being based on Open APIs and architecture. It will allow users flexibility in workload sourcing options without getting locked-in to a particular public cloud or vendor solution. It will enable service providers to rapidly offer a hybrid cloud solution.
We believe that an open and as easy-to-use approach is essential to delivering real hybrid cloud capabilities and help transform the way IT services are delivered. With such an approach; IT will not have to live in the shadows of “Shadow IT” rather can act as a broker of cloud services for LOBs. As the above requirements are addressed, we will increasingly see organizations taking a hybrid approach to cloud.
Cisco’s award winning converged infrastructure management software solution just got even better! In the latest release of Cisco UCS Director, we’ve added broader and deeper infrastructure support across the compute, network, storage and virtualization layers, as well making the product even more scalable. The latest release also comes with new software development kits (SDKs) for partners to provide extensibility and interoperability and a northbound API for integration with Cisco’s Intelligent Automation for Cloud (IAC) as well as other cloud management systems. Let’s take a closer look at these enhancements in detail.
SDKs and APIs
UCS Director’s new SDKs and APIs were recently made available to give our partners the following functionality:
Northbound API which integrates into higher platforms like Cisco IAC. This API enables you to perform operations on Cisco UCS Director resources and to integrate those operations into applications so that they can provide API-supported functionality and features.
SDK available on cisco.com & DevNet (CDN), which provides open automation for partners looking to create new device connector
UCS Manager 2.1 (Delmar); KVM console; SAN zoning through UCS Manager; standalone management of UCS C-SeriesUCS Integration -Local disk provisioning and policies
-Workflow tasks for vNIC templates and service profile
-Assign blade or server pool to group
-Cloning of profile and profile template
Whiptail – application acceleration
Invicta, Acella API Version 1.5 support
CRUD actions, workflows, and reports
Converged and stack views
UCS Central Integration
-Multi-domain Manager concept
-Inventory for UCS-M elements
-Local/global service profile and templates
-Pools and policies for network and storage equivalent to UCS-M
We are almost at the end of January in 2014 and Cisco Live is underway in Europe. The IT landscape is changing rapidly, and we all need to adapt to it. In a previous blog I had listed some of the high priority issues for IT in 2014, and in this blog I will emphasize the rate of change needed to keep pace.
Cyber threats are growing in all forms and shapes.
From retail giants like Target to local school districts, all have been targets of attacks last year. This 2014 Cisco security report is an eye opener. The use of cloud computing and mobile devices is only increasing the complexity of security. Vulnerabilities are increasing because data centers have thus far adopted “bolt on” security measures i.e disparate technologies with multiple control points that were never meant to work together. Malicious actors are working faster to exploit gaps. IT must adapt quickly to this new normal.
IT must now deliver services not technology
Business expects IT to deliver consistent high quality services. Projects will not be one off, siloed, custom solutions. This is akin to moving away from a job shop model to a highly industrialized continuous flow model in the world of manufacturing. Projects will get more complex due to the use of shared resources and the resulting interrelationships. All this calls for a high degree of automation and service assurance mechanisms, resulting in more changes to the IT department. At Cisco, we are working hard to provide IT the necessary tools to adapt to these changes. UCS Manager is still the gold standard in automating converged infrastructure setup and maintenance. This is made possible with the service profile innovation shown in the video below. To find out more about UCS Manager attend this meetup in Santa Clara on the 27th of Feb.
Ciao! The data center and cloud management software team is in Milan this week to showcase our Unified Management solutions in the Data Center and Cloud booth as well as the Cisco ONE Platform booth at our annual Cisco Live customer event in Europe.
So, please check out the live demos, join our breakout sessions, and take this opportunity to how our Cisco’s Unified Management software solutions can help your organization improve efficiency, become more agile, and reduce costs.
See below for details and stay tuned for some exciting announcements …
Cisco and CA Technologies have been solution partners for years integrating CA IT management solutions with Cisco Data Center and cloud infrastructure. On Jan. 27th, at CiscoLive Milan, the two companies will demonstrate the next generation data center – highlighting synergies in DevOps automation and Unified Infrastructure management. Why is this, a turning point in IT evolution?
In a recent blog, I talked about the growing agility gap in mid-to-large data centers -- that in spite of a decade of unified fabrics, virtualized services and controller abstractions, the demands for scale, performance and security are growing faster than IT can satisfy them. This is due to new mobile, social and Big-Data applications that are much more dynamic due to multi-tenancy, higher demand peaks, more distributed users, broader device support, varying performance needs, 24x7 global usage, and changing security vulnerabilities.
I also said that while the evidence of the agility gap is showing up in ballooning network reconfigurations (such as ports, VLANs, subnets, security services, load-balancing, flow optimizations, monitoring), the solution lies in accelerating DevOps – the relatively new discipline of translating dynamic application and tenant needs into an optimal data center topology – and not just in a controller based programming.
CA Technologies is now showing that DevOps processes can be accelerated by orders of magnitude using Cisco Application Centric Infrastructure (ACI). This is resulting in application optimized infrastructure, much lower TCO and faster time to new services.
DevOps, the frontline in closing the agility gap
DevOps evolved from the need of global enterprises to meet competitive application demands with “continuous application delivery.” Financial services, Healthcare, Hospitality and Retail are among the sectors for whom the frequency of application changes have shrunk from 18 months, to daily and even multiple times a day!
Drivers include mobile apps that need continuous enhancements, cloud scaling, Big-Data and the use of Agile Methodology where major code changes occur in weekly or daily “sprints.” DevOps brings continuous integration and operational processes to allow several Developer, Quality Assurance and Application-production teams to collaborate to improve agility of application deployments.