It’s Spring Cleaning Time for Your Data Center

 

Spring may mean warmer weather, but for most of us it also means getting in touch with our inner hoarder: seriously digging into our closets, cupboards and drawers and parting with stuff we don’t use anymore. The alternative, of course, is watching helplessly as our dens, basements, hallways and garages simply become makeshift storage rooms.

As frustrating as spring cleaning can be, it’s a piece of cake compared to the hoarding that’s going on in the average data center. Consider the astronomical growth in data that’s causing organizations’ storage needs to rise by 40% a year – all while IT budgets remain flat and data center resources are stretched to the limit. You can dig into your closet and toss your acid wash jeans from 1993 or that gaudy bowl you got from your aunt in Albuquerque, but how do you toss gigabytes and terabytes of data you can’t see? Where do you start? [Read more…]

To scalability and beyond! Part 2

It’s time to get started

Once your captains er– leaders are on board with the benefits and significant cost savings an efficient data center can deliver, it’s time to assess your current infrastructure and create a plan. The very first thing you need to do is,

 “Shunt the deuterium from the main cryo-pump to the auxiliary tank.

 Just kidding. But there are best practices and procedures available to ensure you have a successful implementation. To get started you can:

  • Complete an energy efficiency assessment
  • Benchmark data center performance
  • Rate data center energy efficiency
  • Look for ‘quick win’ opportunities to improve efficiency (these can add up to massive gains)

You need to be able to show your energy usage before and track improvements over time. This will make a big difference if compliance becomes an issue or new energy incentives come into place. When you’re ready to drill down, take a look at each area of your data center to see where can make room for improvement.

 Servers

Look for old server environments and applications that can [Read more…]

The network balancing act.

Of virtual domains and load balancing: some benefits of network virtualization.

 It’s a common problem: by underutilizing their resources, most data centers end up deploying more servers than they need to. Those additional servers increase costs, of course, but they also create more disparate environments that are tough to manage. But now IT has the capability to buy single device items that, through virtualization, they can carve up into separate logical devices that look physically separated.

 That means you can create a virtual domain for a company or entity or line of business that’s run and managed separately, discreetly from other things in the environment even though each is sharing the same physical device.

 It also means increasing the value of your network devices and leveraging your infrastructure.

 Yet virtualization of data center infrastructure can bring with it an increase in security issues — around privacy, compliance, and business continuity, for example. So a lot of technology is being offered by vendors that can make sure, for instance, that devices coming on the network are trusted and that you don’t end up accidently putting something online or in a place where it shouldn’t be and create security issues.

 Another challenge when it comes to network virtualization centers around [Read more…]

It’s time for a little housekeeping on the data center floor.

What’s Unified Fabric Networking technologies?

 As the number of data, applications and servers in data centers have increased, so have corresponding storage and networks. Couple this with an increasing appetite for Virtualization and networking seems to have been a bit forgotten.  As organizations look to condense platforms and applications the demand on networking and storage connectivity continues to increase.  Most organizations have separate fabrics, and redundant networking switches, for their IP data and SAN connectivity.  This adds additional complexity and equipment in the environment that has to be managed, monitored and controlled.

 The place has gotten crowded to put it mildly and it’s reduced the flexibility of organizations to adapt to changing business needs – like getting onboard the cloud.

 One answer to this housekeeping challenge has been a new standard called Data Center Bridging (DCBx), which collapses IP and Storage networks (regardless or protocol) down to a single common network infrastructure and unifies all these disparate networking fabrics – eliminating a lot of switches, cables and more.

 This converged infrastructure – or Unified Fabric Networking as it’s often called – is really about unifying the fabrics of connectivity on the data center floor so that as much as possible works through one single wire. Now you’re able to provision any connectivity you want across that wire. No longer do you have to bring down a machine and recable or bring in more switches. It can all be done elastically and dynamically. [Read more…]

Good news, bad news and more good news: How virtualization is streamlining branch offices.

They say, be careful what you wish for. A plant needs sunshine and water to grow but give it too much and the plant burns or withers. The same sort of predicament might be said of data center consolidation.

Over the last few decades, pulling the plug on inefficiently used servers in a variety of data centers and branch offices throughout an organization’s network and replacing them with a large number of x86 servers in a smaller number of data centers drove impressive economies of scale. That was the good news. The bad news was that these massive data centers turned out to be more complex to manage, consumed a lot of energy and depended on sophisticated planning to deal with unexpected loads.

That, of course, resulted in more good news in the form of server virtualization to address these challenges. Virtual servers could be provisioned more quickly than physical ones, they needed less space and power and they could be cloned, moved or clustered without service interruption.

 But server virtualization, you guessed it, has led to some challenges of its own, [Read more…]

Virtualization in Linux environments: The Penguin wants to play.

While it’s true almost everyone is gung-ho about the benefits of virtualization these days, many organizations end up missing the boat on maximizing its greatest benefits – increased server utilization and consolidated workloads, lower energy costs, increased flexibility and easier system management.

 Why? Maybe because they view virtualization as a magic bullet to solving all their hardware and computing problems. It’s not. And it turns out whether you’re virtualizing a Linux environment or any other, there are steps to take, best practices to follow and pitfalls to be wary of to ensure virtualization implementation and ongoing management run smoothly.

 Linux.com’s Joe ‘Zonker’ Brockmeier offers a wealth of common sense suggestions –ones that are often overlooked or taken for granted – here and here to help you virtualize successfully in Linux environments:

 Define your goal, perform and inventory and set up a roadmap: Understand what you want to accomplish, have a well-defined set of goals, identify hardware that will be freed up or phased out, and come up with a detailed requirements document that outlines the hardware you’ll need, as well as storage, management and possible solutions. A lot of this may elicit a ‘duh,’ but issues later can often be traced back to insufficient planning and goal-setting early on.

 Beware of “virtual sprawl”: Just because it seems easy to deploy a virtual machine doesn’t mean you should. “It’s important to manage virtual machines as if they were [Read more…]