Infrastructure Management Software Blog : server virtualization

Operational Excellence - Article on Mark Hurd

pspielvogel — Mon, 02 Mar 2009 18:59:00 GMT

There was a recent article on Mark Hurd in Fortune Magazine. It focuses on his obsession with numbers and execution.

While the article does not mention infrastructure management software, readers of this blog will immediately recognize how important consolidated event management, server virtualization, and automation are to reducing operating expenses.

For Operations Center, Peter Spielvogel

Virtualization: IT Nirvana or CapEx for OpEx Tradeoff?

pspielvogel — Mon, 09 Feb 2009 17:34:00 GMT

Today’s post is from Kalyan Ramanathan, who manages virtualization initiatives for our data center automation group. This is the first in a series of posts on how organizations use automation to improve their IT infrastructure management.
- Peter

Even in this down economy, data center virtualization is all the rage and that is for a good reason - virtualization brings many benefits, including infrastructure consolidation, reduced power usage, data center footprint etc. Not everything is rosy in this picture, though. Virtualization also introduces several challenges and IT management is definitely a critical one.

The beauty of virtualization is that this dirty little “management” monster only rears its head once you are deep into the dungeon!

Let us explain what I mean in the context of a recent conversation that I had with the data center managers of a large financial enterprise. This enterprise started out with a consolidation project and instantly found value by virtualizing its servers. By consolidating its Windows and Linux servers, IT was able to see a 20-30% reduction in servers counts. Fewer servers meant lower power usage and freed up data center racks. Managing the initial virtualization candidates – print server, Web servers was not that bad either, and a basic set of scripts and tools sufficed. The CIO is thrilled with the instant CapEx saving and IT became a hero.

Fast forward a year.

With their initial success, data center managers pushed to virtualize a large portion of their data center. So now that same IT department has 3x the number of servers. IT has more vendor technologies to contend with – VMware, Microsoft HyperV, Xen, etc. and each of these servers – hypervisors and virtual machines - needs to be configured and patched in a economy that prevents any IT headcount increases. And what about the agility and flexibility benefits?

While managing the homogenous Web farm servers was simple, the same cannot be said of complex applications that depend on resources such as storage, network etc. The data center managers were facing a severe operational issue. Left unattended, this problem was only going to become worse. “We traded CapEx for OpEx”, noted the director of operations, as he articulated their management challenges.

Are you seeing similar problems in your datacenter? Post a comment about your key management challenges.

You might want to read a free e-book by Realtime Publishers on Virtualization and Service Automation for some shortcuts on how to start a virtualization initiative and several pitfalls to avoid.

For Business Service Automation, Kalyan Ramanathan

Making Infrastructure Management Less Taxing

pspielvogel — Tue, 20 Jan 2009 00:29:00 GMT

I bought my copy of TurboTax over the weekend. While waiting in the long checkout line, it occurred to me that the United States tax code has several similarities to IT infrastructure management. Let me explain.

At the beginning, there was a needs assessment, planning process, growth forecast, and then implementation. When the system went live, everything was simple and all the pieces worked smoothly together. Over time, came growth, “enhancements,” complexity, and some unintended consequences. Current state of affairs: a mess. But at least we know the system’s idiosyncrasies or at least have processes in place to make everything work.

Quick, which am I talking about, your IT infrastructure of the tax code?

In a new datacenter, companies design in the management tools that will help them best meet their end-users’ needs and run the equipment to gain maximum efficiencies. Over time, requirements evolve (expand!) and new technologies such as virtualization proliferate. IT managers keep their environments current by adding new management tools, often with new software focused on the most recent infrastructure enhancement.

Modern datacenters include dedicated tools for their latest rack of blade servers that include environmental and power monitoring, software to dynamically move virtual machines from one physical server to another without the users even noticing, and monitoring consoles that give you metrics based on the actual performance your end-users experience.

But, even with all these innovations, it’s still really hard to keep track of all the moving parts in any IT infrastructure. So, periodically, companies undergo datacenter consolidations to streamline their operations, reduce the number of management tools, and make the whole process less taxing.

If you want help in simplifying the tax code, contact your local member of Congress.

If you want to simplify your IT infrastructure management, HP has a variety of solutions to help you. We’ll talk in depth about datacenter consolidation and server virtualization in future posts.

For HP Operations Center, Peter Spielvogel.

Does Virtualization Consolidate Your IT Operations Or Fragment Them?

pspielvogel — Thu, 15 Jan 2009 02:06:00 GMT

Customers tell me constantly they are consolidating their tools sets to drive efficiency and reduce operating costs. They tell me they do not want hundreds of point tools that they have to integrate themselves. They tell me they want to spend more time working on strategic projects and less on keeping the lights on. They tell me that they have Virtualization projects that are following the same path as previous innovation projects. When is it going to stop?

The average IT organization spends 70-80% of their budget “keep the lights on” and not innovating. As a result, most IT organizations are looking for ways to reduce their IT operations costs so they can free up resources to work on more strategic business demands.

Most customers are modifying their cost structure through consolidation projects (Data Center Transformation). The goal for these consolidation projects is to eliminate redundancy and more importantly increase efficiency to avoid and minimize downtime.

There are many types of IT consolidation that can occur but the most common are organizational, and datacenter consolidations. In both cases virtualization is often one of the key enabling technologies deployed. Each of these projects usually kicks off an IT management tools re-evaluation.

Unfortunately as IT organizations invest in these projects they are often very disruptive and counter productive to IT initiatives focused on reducing operating costs. I believe it is due to the lack of planning on how to best manage the new technologies once deployed.

Virtualization promises to deliver significant hardware and software license cost savings and thus very attractive to companies trying to reduce capital expenses. What is not understood by most companies is the fact that every new technology deployed can be very disruptive to IT Operations goal of keeping the lights on and potentially increase operating expenses.

Virtualization is not alone. Think back when your organization first deployed new technologies like, PCs, Client Server, Web, ERP, SOA, etc. How disruptive was it? How long did it take to get the management technologies embedded into your IT Operations processes correctly? For most companies it was too long, which drove the need to consolidate the number of management vendors because it was too costly and inefficient to maintain multiple home grown integrations, event management systems, dashboards, and IT processes.

Most of the disruptions were because IT organizations do not know where or how to integrate the data received by the multiple tools deployed or where different organizational boundaries should lie. This is where HP Software and Solutions has helped thousands of customers over the past 15 years drive the cost out of IT operations through tools consolidation projects.

What disruptions have you encountered during your data center consolidation project? Or, how did careful planning avoid the pitfalls that trip up most IT professionals.

For HP Operations Center, Dennis Corning

Is Managing Virtual Servers Just Like Managing Physical Servers?

pspielvogel — Fri, 09 Jan 2009 23:30:00 GMT

As I travel and speak with customers, partners and analysts firms, I keep hearing that ownership for virtualization management is fragmented within the IT organization. Some companies have even created a new role called the Virtualization Administrator. Why?

If virtual servers are just like the legacy “physical” or dedicated servers why is there such a fragmentation in monitoring responsibilities, tools and processes today? Haven’t we learned from previous technology innovation that when it comes to IT Operations, the sooner the monitoring tools and process are streamlined into the standard operating procedure the better?

Take a look inside your organization today. Are you fragmenting the monitoring and management of your Virtual Servers? Are you deploying point products to manage your Virtual Server farms separate from your physical servers? You know from previous projects the sooner you streamline the management tools into the process the more efficiently IT can keep the lights on while freeing up resources to execute more strategic projects.

So what do you need to successfully manage Virtual Servers, the applications running on them and the underlying network? The same tools you already use to manage these entities prior to virtualization (HP Operations Center, formerly OpenView), with some enhancements to monitor the virtual layer.

The enhancement comes in the form of a new HP Smart Plug In (SPIs) for Virtualization. This new SPI for Virtualization plugs into HP Operations Manager and delivers the following:

Uses a common methodology to monitor both Physical (dedicated) and Virtual Servers.
Monitors both performance and availability of VM host and guests independent of each other to provide current status
Automates the creation of baselines and setting thresholds to reduces configuration overhead
Auto-discovery and visualization of the virtualized environment—including the dependency between the ESX host system and each guest system and updates service navigator views
Provides a single console to simplify operation tasks. When a metric exceeds a defined threshold it alerts Operators and provides instructions for fixing the problem and/or launch corrective actions automatically or manually
Delivers comprehensive reporting on the short- and long-time behavior for a virtualized system.

The alternative is to use a native tool that came with your virtualization platform or buy a niche point product. Both approaches can be effective for the Virtualization Administration team but does it enable you to drive the cost out of IT Operations? My recommendation is to extend your HP Operations Center deployment to ensure you continue to optimize your cost structure.

Let me know why or why not centralizing your physical and virtual operations into a single console is right for you.

For HP Operations Center, Dennis Corning