CloudBolt Blog

In numerous conversations with customers and analysts it has become clear that a consensus across the industry is that Cloud Managers are as game changing for IT as server and network virtualization themselves.  Among those looking longer term at the promise of Cloud Computing (Public, Private and Hybrid), it is clear that the Cloud Manager will become the keystone of value.  Many people’s opinion is that Cloud Managers are the initiator of next major wave of change in IT environments.  How?  Well let’s look to the past to predict the future.

Proprietary Everything

Back in the early 80’s, general purpose computers were first spreading across the business environment. These systems were in the form of fully-proprietary Mainframes, Minicomputers. The hardware (CPU, Memory, Storage, etc.), Operating Systems and any even any available software were all from the specific computer manufacturer (vendors included DEC, Prime, Harris, IBM, HP, DG amongst others).  Businesses couldn’t even acquire compilers for their systems from a third party.  They were only available from the system’s manufacturer.

Commodity OS leads to Commodity Hardware

The advent of broad interest and adoption of Unix started a sea change in the IT world.  As more hardware vendors supported Unix it became easier to migrate from one vendor’s system to another.  Additionally, vendors began building their systems based on commodity x86-compatible microprocessors as opposed to building proprietary CPU architectures optimized around their proprietary OS.

Architecture-compatible hardware not only accelerated the move to commodity OS (Unix, Linux and Windows), but in turn, increased pressure on vendors to fully commoditize server hardware.  The resulting commoditization of hardware systems steeply drove down prices.  To this day, server hardware largely remains a commodity.

Virtualization Commoditizes Servers

Despite less expensive commodity operating systems and commodity hardware, modernizing enterprise IT organizations were still spending large sums on new server hardware in order to accommodate the rapidly growing demand of new applications.  In large part, IT organizations had a problem taking full advantage of the hardware resources they are spending on.  Server utilization become a real issue.  Procurement of servers still took a considerable amount of time due to organizational processes.  Every new server required a significant amount of effort to purchase, rack and stack, and eventually deploy.  Power and cooling requirements became a significant concern.  The integration of storage, networking, and software deployment and maintenance still caused considerable delays into workflows that are reliant on new hardware systems.

Server virtualization arrives commercially in the late 1990’s and starts getting considerable traction in the mid 2000’s.  Virtualization of the underlying physical hardware provides an answer to the thorny utilization issue by enabling multiple individual server workloads that have low individual utilization to be consolidated on a single physical server.  Virtualization also provides a limited  solution for the  the procurement problem, and helps with the power and cooling issues posed by rampant hardware server growth. Areas of networking, storage, and application management remain disjointed, and typically still require similar times to effectively implement as before the advent of virtualization thus becoming a major impediment to flexibility in the enterprise IT shops.

Now we find ourselves in 2013.  Most enterprise IT shops have implemented some level of virtualization. All of the SaaS and Cloud-based service providers have standardized on virtualization. Virtual servers can be created rapidly and at no perceived cost other than associated licenses, so VM Servers are essentially a commodity, although the market share for the underlying (enabling) technology is clearly in VMware’s favor at this point.

The problem with these commodity VM servers is that making them fully available for use still hinges on integrating them with other parts of the IT environment that are far from commodity and complex to configure.  The VM’s dependency on network, automation tools, storage, etc. hinder the speed and flexibility of the IT group to configure and provide rapid access to these resources for the business.

Network Virtualization arrives

A huge pain point in flexibly deploying applications and workloads is the result of networking technology still being largely based on the physical configuration of network hardware devices across the enterprise. The typical enterprise network is both complex and fragile, which is a condition that dos not encourage rapid change in the network layer to accommodate business or mission application requirements. An inflexible network which is available is always preferred to a network that failed because of unintended consequences of a configuration change.

In much the same way as Server Virtualization abstracted the server from the underlying hardware, Network virtualization completely abstracts the logical network from the physical network.  Using network virtualization it is now possible to free the network configuration from the physical devices, enabling rapid deployment of new, and more efficient management of existing virtual networks.  Rapid adoption of network virtualization technology in the future is all but guaranteed.

Commoditizing all IT resources and compute

With both network and server virtualization, we are closer than ever to the real benefit of 'Cloud Computing': the promise of  fully commoditized IT resources and compute.  To get there, however, we need to coordinate and abstract the management and control the modern enterprises’ internal IT resources and compute resources being consumed in external public cloud providers.

To enable rapid and flexible coordination of the IT resources, the management of those enterprise application resources must be abstracted from the underlying tools.  The specific technologies (server virt, network virt, automation, storage, public cloud provider, etc.) involved are viewed as commodity, and can be exchanged or deprecated without negatively affecting the business capabilities of the enterprise IT. Additionally this abstraction allows the IT organization to flexibly adopt new and emerging technologies to add functionality and capability without exposing the business to the often sharp edges leading edge technology.

The necessary resource abstraction and control is the domain of the not just the virtualization manager-- but really the Cloud Manager. In short, the Cloud Manager commoditizes compute by commoditizing the IT resources across the enterprise and beyond.

With such an important role it is no wonder that every vendor wants to pitch a solution in this space. The orientation or bias of the various vendors’ approaches in developing a Cloud Manager for enterprise IT will play a critical role in the ultimate success of the products and customers that implement them.

Over at Information Week, Jim Ditmore discusses how advances in CPU power efficiency will eventually save businesses significant data center costs.

It’s certainly a compelling case, but there’s an assumption being made—that VM count will not grow at the same pace as the gains in hardware efficiency.

Many customers I speak with are certainly excited about the prospects of more efficient data centers, both in terms of CPU performance and efficiency.  One common problem they’re butting up against, however, is that of VM sprawl. Unused or under-utilized VMs in an environment have a significant impact on the overall efficiency numbers that an IT organization can expect to see.  If VM count increases at the same rate as the processor/efficiency curve, then the net result will be as it is now: the amount of required hardware to sustain the existing load will increase. 

To his credit, Jim comes close to calling this point out: 

“You'll have to employ best practices in capacity and performance management to get the most from your server and storage pools, but the long-term payoff is big. If you don't leverage these technologies and approaches, your future is the red and purple lines on the chart: ever-rising compute and data center costs over the coming years.”

Efficiency doesn't matter when VM Sprawl consumes additional capacity provided by more powerful and efficient CPUs.

But that’s still making the assumption that It organizations are well-prepared to effectively solve the issue of VM sprawl.  For many of the customers I work with, that’s a pretty big assumption.  IT Organizations are well aware of the impact of sprawl, but have few tools to effectively combat it in a reliable and consistent manner.  Additionally, the sustained effort required to maintain a neat-and-tidy virtualization environment (at least regarding sprawl) is often great, placing much pressure on an IT organization that’s likely already seen as lacking agility and responsiveness to the business.

The default solution, which of course is rife with issues, to this struggle is well known, and relatively easy:  Throw more hardware at the problem.  Or push workloads to the public cloud.  Either way, it’s just a Band-Aid, at best and does nothing to contain costs into the future.

The only way for IT organizations to benefit in an ongoing basis from the processor performance/efficiency curve is to effectively control sprawl in the virtual environment.

And how does one do that?  With a Cloud Manager like CloudBolt C2.

I provided some background into how cloud models—be they public or private—break both traditional software licensing models, but in some cases, the software licenses themselves.  Despite the growing pervasiveness of cloud-based compute, I was amazed that there are not just cloud-incompatible license agreements, but also some licenses actually prohibit the running of certain software in the cloud.

There’s a term that’s applicable here: BYOSL (Bring Your Own Software License).  Vendors who understand a cloud-centric IT model understand that you need to be able to use their software in cloud-centric models if that’s what your organization wants to do.  However, many large vendors (it won’t be hard to find which ones I’m talking about here) take specific steps to limit BYOSL (just to name a few I came across):

• Require specific understanding of underlying hardware architectures or processor specifications
• Require licensing based on the physical, not virtual host
• Mandate customers run vendor-provided license tracking, further complicating multi-location or multi-environment installations
• Prohibit software from being virtualized
• Force purchase of higher-cost public cloud resources which rope in the underlying OS license regardless of customer license availability
• Force purchase of “License Mobility” options in order to run software in public clouds

In short, software license management in the cloud is a mess.  What’s really odd is that those large vendors also claim to know about cloud.  My resulting questions to you are simple:

• Are you relying on cloud strategy from a company that actively uses their software licensing to discourage or prevent you from moving to a more open cloud-centric IT model?
•  Are they leveraging their licenses to force you down the path they want you to?

My subtext:  If you’re listening to intently to those large vendors, the answer to both is “yes”.  Proceed with caution if your primary cloud strategy comes from your hardware, middleware, database, or even OS vendor.

Be sure to get the full story from vendors and how software licenses will work in the cloud.

In some cases, I can understand why these companies believe that licensing software based on the type of physical CPU, or the count of virtual CPUs, or even the type of cloud the application is running in.  However, more often than not, vendors are just trying to protect market share or revenue.  In other situations, perhaps there are other requirements, such as performance, or hardware that present troubles when run virtualized.  But that’s a support issue, not a licensing one.  Otherwise, their restrictions make little sense.

Vendor licenses that restrict where purchased software can and cannot run are incompatible with the current trajectory of IT.  Vendors need to understand that there is little fundamental difference between applications running on an Amazon EC2 instance in Virginia vs. a Google Compute instance in California, vs. a Rackspace instance in Texas vs. your own VMware-based vCenter cluster in your data center.  

Cloud management can help to solve this problem, too.  If software vendors are concerned about protecting revenue, they have no impetus to alter a license model in a way that will save customers money but be totally untraceable.  IT is absolutely heading toward usage-based accounting and costing of license usage, but currently, vendors possess no reliable way to effectively track that usage.  Cloud complicates that for them.  A vendor-agnostic tool that delivers accurate license usage counts in real-time ensures IT organizations remain compliant, and gives software vendors a little peace that they’re not missing out on revenue, but also provides IT organizations the tool necessary to much more effectively utilize software licenses. 

And for those software vendors that are behind the times?  They must adapt, or they’ll be (or rather, continue to be) pummeled out of existence.

Call it what you'd like: Cloud Management Platform, Cloud Manager, or even a Virtualization Manager, but one fact remains-- technology that unifies the management and provisioning of your IT environment is here to stay.  If you're not already looking at one to help reduce I&O expenses, you should be. 

At a high level, Cloud Management Platforms such as CloudBolt Software's Command and Control (C2) provide a layer of management on top of existing virtual environments.  We drive a greater value from existing virtualization and data enter automation/configuration management tools by unifying the management and visibility.  A world-class Cloud Management Platform provides visibility and control of your IT environment mapped to your business via a single pane-of-glass.  That sort of cohesive internal management makes IT more agile and can lower costs.

Of course, we can't stop there.  In a recent Gartner poll, 47% of IT groups planned to have some type of hybrid cloud deployment by 2015-- so any reasonably capable Cloud Management Platform must also provide the same level of management and oversight over public cloud instances and deployments.

Together, these capabilities allow IT line of business owners to rapidly transform their organization into a more agile part service-provider and part cloud broker.

This, according to Gartner, is the likely path of successful IT organizations that are under a near-constant assault from public cloud providers such as Amazon Web Services and Rackspace.  There is no area of business IT that is safe from this attack.  Public cloud vendors in the US and other countries are even now racing to stand up government-only environments, which meet the copious requirements—which often amount to pages of standards required to host many government applications.

Transformation must happen.

To me, the conclusion for any business leader in IT seems to be crystal clear:  transform or be transformed.  This is where effective cloud management comes into play.  By all reasonable accounts, you're either already virtualizing, or have virtualized most workloads capable of being hosted in this model.  The task does not end there.  You must transform your IT organization to be more agile-- more able to respond to the needs of the greater business in less time, and lower cost than the big public providers are able to accomplish.  The only path to this situation is effective cloud management.  

It's a little more complicated than just providing a self-service portal to your users.  There is still significant amounts of policy that need to be updated and adapted to a cloud-enteric world, be it private, public, or hybrid.  You need to continue to provide end-to-end life cycle management, and must be able to deliver the applications that end users need, while doing so in a manner that is controlled and well understood.  

Our recommendation?  Layer in a cloud management technology that will work with what you have in place today.  This way, you have the ability to adopt new technology and adjust your processes gradually, and as it makes sense, vs. the infrequently successful and high cost approach of large infrastructure deployment and vendor lock-in.  Gradually molding your business processes into a cloud-centric model produces a higher success rate, as it provides you the ability to gain immediate benefit: getting resources to users more rapidly, but while also allowing you to plan how best to further implement your newfound flexibility.  This approach solves many common needs:

• Move dev/test to public cloud resources?  
• Implement resource quotas? Modify approval processes?
• Make more public cloud resources available as needed?
• Implement charge or show-back accounting for resource consumption?   

Have a look at our Cloud Management Platform.  We think you'll find it not only has industry-leading capabilities, but will be easier to use and integrate into your existing processes than literally anything else out there, and at a CapEx and OpEx that cannot be beat by anyone in the industry.  But don't take our word for it-- download it for free and get started today.

Proprietary hardware and processor architectures have been outpaced and eventually replaced by commoditized Intel and AMD x86 and x86_64 hardware platforms. Along the same lines, Linux and other open-source operating systems fully commoditized the operating system, ensuring that the underlying architecture really isn’t too important anymore. The next logical step in this was to further commoditize the operating system and hardware platform together—which is what server virtualization does. Server virtualization makes the hardware the ultimate commodity. Often the management layers of the best virtualization platforms are intelligent enough to even appropriately handle differing processor specifications and memory configurations on the hypervisors.

The Intel-based x86 chip revolutionized IT.

A fully server-virtualized environment, however, is still reliant on several things:

1. The administrators are still required to know and understand where and what is being deployed virtually. 
2. When a user makes a request for a resource, it’s got to be put somewhere, and that underlying virtualization technology is something that has to be dealt with, understood, and eventually manipulated in a manner to deploy the requested resource. 
3. The idea of IaaS and PaaS disrupts this a fair amount, but there’s still a choice—a implicit understanding that your requested compute resource is dependent on a single underlying technology, be it from EC2, Google Compute Engine, VMware, RHEV, Xen, Hyper-V, or anything else.

The next step in organizational IT maturity has to be the full commoditization of that compute layer. Just as organizations can now procure commodity servers and storage from a variety of vendors, and abstract that hardware choice using virtualization, so must the actual virtualization technologies be abstracted from the end user. In the end, this makes sense. Users don’t need to know or care where their compute is coming from. They just want access to the resources and services they’ve requested when they requested them. Just as administrators have the ability to choose amongst various hardware providers without affecting users, they should be able to choose amongst differing physical locations, virtualizations technologies, and even cloud providers.

This is where CloudBolt steps in. CloudBolt C2 commoditizes the compute layer. Regardless of the virtualization or cloud technology present, CloudBolt C2 makes compute resources available to users. This is regardless of what the underlying virtualization technology is, where the resources are located, and increasingly, without concern of what that underlying hardware architecture is.

A commoditized compute layer is interesting, but when coupled with end-user self-service, an IT organization has the ability to introduce a tremendous amount of IT agility into the organization.

Give us a call and let’s chat about where your organization is along the path of providing automated self-service infrastructure, applications, and services to your users. We’ll show you how CloudBolt C2 can revolutionize how you look at compute and manage your resources, wherever they are located.

When I first started at Red Hat, I was a consultant who helped customers deploy a relatively wide set of technologies. A common delivery that we performed frequently was to install and configure the Red Hat Network Satellite Server, which did a phenomenal job of deploying and managing its namesake: Red Hat Enterprise Linux. Managing resources is fundamental to operations in a data center, and within a short time, a new breed of management tools started to materialize. These tools promised—and many delivered—one tool with a single-pane-of-glass actionable management of every server resource in a business datacenter, as long as you only tried to use it for the products those tools managed, that is.

The technologist in me was instantly skeptical of single tool products. As I saw it, the problem was more nuanced than “one tool to rule them all,” because most IT shops had more than just a single platform in play. Red Hat Network Satellite server could do things to and with RHEL systems that these other tools couldn’t touch, but it couldn’t do anything on Windows. In any case, why would you want that? You probably had Microsoft System Center to handle your Windows. In fact, most major enterprise platforms have a powerful management tool built alongside it, and also have the ability to control even the smallest configuration. How can a “we-do-everything tool” ever hope to match the power already built into these products for management and configuration?

It’s no different with cloud management platforms (CMP) - they run into the same walls, and don’t have the ability to talk to anything except their own product line. In order for a CMP to qualify as a true “one tool” management product, it has to talk to everything in the datacenter across physical and virtual platforms and cloud frameworks, and help you make use what you’ve already invested in, or it is just another tool that adds weight and costs to a datacenter. And, a “one tool” product shouldn’t require the replacement of expensive technologies or require a significant investment in time and money to install, use and maintain, either.

I would put my skepticism to rest if there was a tool that could give you situational awareness of your entire IT environment while still being able to use the deeper functionality of the native management utility in a manner that leverages rather than replaces your existing tools. This means “one tool” that can identify what is in use, talk to it, allow for full utilization of its capabilities and, perhaps even as icing on the cake, provide it to end-users. Furthermore, the approach must salvage the significant investment in acquisition, implementation, and training you’ve made in those other tools without ripping or replacing anything. Why should a company have to start over, or spend a huge amount of money to buy new stuff so they can use what they already own?

Do-it-all technologies may be very good at certain tasks, but most will fall down when it comes creating a resource management tool that works across the entirety of the resource pool and cloud frameworks, uses existing assets and doesn’t require any significant amount of time or money to implement. At CloudBolt, we’ve focused all of our efforts on a full-on solution for the problems virtualized environments face with a product that can evolve with a company as it makes changes in IT, even as it solves the challenges of right now across multiple platforms and cloud frameworks. Our development team, under the leadership of our CTO Bernard Sanders, and lead product architect Auggy Da Rocha, struck out to create a “one tool” resource management product that can truly rule them all by solving all of those problems, and making maximum use of the capabilities of existing IT as it deploys a private cloud with hybrid cloud functionality. A huge part of my job here is to make sure our customers are getting solutions that provide them with full enterprise situational awareness as well as interoperability with existing management tools. CloudBolt C2 is one tool to rule them all, and it can truly revolutionize how organizations work with, and use their IT.

An affordable tool that builds on top of what you’re already doing, C2 makes what you’ve got visible and presents it as cloud, so your IT shop gains the ability to unify management across a wide variety of technologies in a way which allows for the easy presentation of any IT resource to end users. Because those users are fully abstracted from the underlying technology, you’re suddenly able to choose and configure the best underlying technology combination possible. Not only do you gain agility in serving end users, you also gain flexibility in your choice of technology, and that turns into the ability to drive down costs on things like VM sprawl. Plus, it decreases possible attack vectors.

By leveraging the existing technologies already in your environment, CloudBolt C2 allows IT shops to realize the full value and potential of existing management and virtualization tools, and gives higher ups a much clearer, fully-transparent view into cross-environment utilization in ways that were absolutely impossible before. I could go on, but you get the idea. In short, using C2 as one tool to rule them all gives you the ability to unlock the true power of your entire IT environment in ways that will save you time, money, and ensure that you leave your end users smiling.