CA Data Center Automation Blog

The industry has seen an upward trend in server virtualization for the distributed environments.  Although virtualization is not new, (it’s been in the mainframe world for the more than a decade) it has gained substantial momentum primarily as a means to support data center management and operational efforts.  A recent research study sponsored by CA shows that the top three drivers behind virtualization are server consolidation, easier hardware provisioning and software deployment, and lower total cost of ownership.  Speaking with customers at a recent IT Executive’s Club meeting in Chicago, where the theme was virtualization, many of the larger shops were using virtualized server environments to reduce both power consumption and physical space (server consolidation) within the data center.  I see this as a common theme across data center environments, also contributing to a “green IT” initiative as a positive side effect (see my previous BLOG, “Data Center Automation and Green IT - Part 1”).

Virtualization promises many advantages in data center manageability including operational efficiency, but it also brings along an additional element of complexity.  This, in fact, is where automation plays a key role in keeping virtualization management in-check with the rest of the data center management strategy. Today, virtualization is a key driver for many data center automation strategies especially as data center managers are faced with better ways to manage virtual assets in the same context as physical asset. For those who already have a virtualization strategy in place, now may be a good time to consider extending the benefits of data center automation, reaping all of the cost benefits and efficiencies that automation has proven to provide. My other DCA Blogs go on to discuss many points of entry into an automation strategy as well as other considerations when deciding to undergo automation.

Given all the attention and interest behind "Green IT", I'm sure this will be first of many Blogs I will write on this subject.  I've held off on writing about Green IT until now, simply because I needed time to consume and digest where Green IT initiatives best fit into a data center automation initiative.  The driving forces behind Green IT are pretty easy to understand, data center and the IT industry as a whole are recognizing the impact of energy consumption, or the overall "carbon footprint" on the environment.  This realization has led the IT sector to search for "green" solutions.  Quite a few evangelists and vendors alike have jumped onto the Green IT bandwagon and therefore have created a lot of attention to the subject.  Many data center automation initiatives already go hand-in-hand with developing and/or delivering green solutions.  Take for example the rate of server consolidation and virtualization, a recent CA sponsored global survey on virtualization indicates that a large majority of US respondents (68%) have adopted a virtualization strategy in order to consolidate servers.  The drive to consolidate promises to reduce physical equipment overhead, power consumption and management complexity.

Server utilization, more specifically "under utilization", also plays a big role in reducing data center power consumption.  Having the ability to run servers at more optimal utilization levels, i.e. 40%-60%, as compared to industry averages of <10%.  This means taking advantage of available server capacity and "automatically" provisioning additional capacity when needed.  Those servers that are not required to be online to meet demand at the time, are either powered down into "cold standby" mode, or in "hot standby" ready to accept demand at a moments notice.  The IT industry is HOT on Green IT and it only makes good business sense.  Reduced power consumption and being a good "corporate citizen" is both fiscally and environmentally responsible and should be part of our civic duty.

As you can see, policy driven automation and virtualization play a key role in meeting Green IT objectives and should be considered as part of any Green IT initiative within the data center.  In future Blogs I plan on exploring other aspects of achieving Green IT goals, such as some of the underlying risks in the goal to become Green.

 

When talking about automation in the data center we think about the use of tools and processes to coordinate and execute on activities with hopes to reduce complexity, errors, labor, and costs while increasing productivity.  To further automation goals, we have to consider outside influences to the data center as well as existing software tools and processes that are already in place.  It is likely that any given data center environment consists of a spattering of technologies across multiple platforms and vendors.  This platform/vendor mix results in increased data center complexity and can best be dealt with by the proper tools for integrating and orchestrating IT processes across IT silos, regardless of the vendor or vendor specific platforms.  Integration is the key concept and Run Book Automation (RBA), aka IT Process Automation (ITPA), is a critical component in any automation initiative.

Recently CA announced an OEM agreement with Opalis, a leader in RBA tools.  This partnership will allow CA to take advantage of an already established, industry leading technology as well as to further enhance its integration points for CA software.  But, this does not preclude the existing integration points across other vendor platforms, nor Opalis' continued R&D efforts as a vendor neutral solution.

So what does RBA bring to the table?  "IT process automation provides the ability to launch a process in context and pass information from one process to the next with a level of accuracy far superior to that of any entry into an administrator interface. Solutions in this space replace the scripting of application production rules (run book)."  The demand for process automation is driven from senior IT leadership looking to: Increase IT operations efficiencies, especially around the adoption of best practices, increase IT agility and proving IT operations' accountability to the business.  

You may have noticed a theme of process automation, coupled with best practices?  Yes, RBA/ITPA can help achieve higher levels of process maturity, notably those specified in best practices frameworks, such as ITIL.  IT process automation can easily assist IT Operations in automating those processes that are established and repeatable, while setting the stage for future process design and improvements.  As IT Operations continues to identify, define and improve on IT processes, ITPA serves as a fundamental step in furthering their goals.

Data center automation (DCA) is designed to deliver value back to the business through increased levels of service and flexibility to meet business needs and demands.  With increasing complexity within the data center, including but not limited to: a growing number of applications, bare metal and virtual server sprawl, security and regulatory requirements, there is an increased need for effective management in minimizing its effects.  
In a recent independent study on data center automation sponsored by CA (http://www.ca.com/dca/survey), the majority (46%) of the respondents identified business compliance/ uptime as the number one challenge that they hope to address with automation. However, the survey respondents also indicated that those who have already implemented some sort of automation (an average of 48% of their data center tasks) saw that only 12% had seen these efforts as highly effective.
This brings into question where to begin an automation effort?"  The right tools are essential in obtaining your automation goals, and tools that are flexible enough support your organization's unique processes and capabilities are an absolute necessity.  Begin by identifying and defining your current and repeatable processes.  This task could present some difficulty depending on where you are in establishing a best practices approach to IT service management (ITSM).  Of course the IT Infrastructure Library (ITIL), takes the lead as an approach to define and map these processes as part of the overall IT services lifecycle.  
In the event that you have not introduced any of these process frameworks or best practices in your environment, simply identify the existing approaches to the routine tasks currently undertaken in the data center such as patch management, maintaining consistency through gold standards, workload automation and/or server provisioning.  Once your routine and repeatable processes have been successfully identified and automated, be sure to continuously measure their effectiveness and implement any changes as needed for improvement.  
Next in your quest for automation, begin to identify and define more complex processes that can bring additional value from IT services to the business.  At this point you may have already experienced a level of "enlightenment" in the quest for data center automation, meaning that you have already freed-up resources (both human and technology), allowing more time to focus on more strategic efforts.  Examples of more complex processes include intricate workloads and policy based provisioning (based on business policy).
Data center automation has many points of entry and is only limited by your organization's ability to plan and execute where it makes sense to the business (cost vs. value).   Take a careful look at what your situation has to offer and evaluate the best course of action for automation in your unique environment.
 

The day-to-day challenges of managing the IT data center include management of its diverse and complex makeup of key systems to meet the needs of its users, but more importantly the overall business.  As IT becomes more aligned with the business, IT begins to lose its longstanding identity as a "cost center" as it is required to provide more tangible value back to the business.  One of the largest hurdles in this quest is to turn the tables on data center management as a "lights-on" cost center to a provider of business value.  We've heard the statistic that 70% of IT's budget is spent on running day-to-day operations - simply keeping the light's on - leaving only 30% of the budget for deploying innovation along with new applications to fuel business growth.  This statistic leads to an interesting question, where in the datacenter do these costs exist? Expensive staffing, high cost of hardware / software, Inefficient processes?  The answer is yes to all the above. And a combination of all the above is easily attributed to overall data center complexity.  So how can we reduce those complexities so that we direct more resources towards providing business value?

Let's begin to dissect data center complexity in order to better understand where the problems lie.  In the perfect world, data centers would have consistency across platforms (hardware, software, OS), staff skills sets and operating environments to support a stable business environment.  But in the real world complexity exists and rears its ugly head, especially in attempts to break these barriers to support a dynamic and constantly changing business requirements.  One step towards datacenter flexibility is the ability to de-compartmentalize and remove the silos of applications, data and operating systems environments yielding positive results in manageability leading to reduction of complexity.  In addition, a consistent and centralized view across these "IT Services" enables IT to invoke higher levels manageability and automation.

I am often asked "is automation realistic for our company? Where do we start?"  Yes it's realistic and one place to start is to phase in Data Center Automation to simply allow IT to free up existing resources from error-prone, tedious and time-consuming tasks for new projects aligned with business strategy. Another may be to enable automation to dynamically provision capacity to meet increased demands for a more stable and flexible environment. These examples show clear benefits in automating processes that normally could take days, if not weeks, to as little as minutes while drastically reducing errors and increasing productivity.  Multiple entry points for Data Center Automation allow organizations to take advantage of the benefits of automation no matter where they are in their process maturity, as well as benefiting from increased process efficiencies as a natural automation byproduct!