open
  1 (866) 866-2320 Resources Events Blog

Challenges Facing Data Centers Going Virtual

Blog

Challenges Facing Data Centers Going Virtual


 

One Thing That Remains Constant is Change.

Technology has expanded by leaps and bounds in the past decade and virtualization is, possibly, the latest fruit from this ever-growing tree. The concept of virtualization has forever changed the processes and applications related to IT infrastructure. 

With virtualization, data centers consolidate their hardware investments, and use their server hardware more efficiently,  expanding  application requirements through virtual servers.

According to the 2010 Data Center Purchasing Survey report, published by SearchDataCenter.com in July 2010, virtualization deployments were expected to continue to increase in the near future. According to the report, approximately 60% of the data centers are planning to expand their virtual network. Virtualization will also be adopted as a secured mode for disaster recovery and resource allocation.  

Data centers can now reach for greater cost-efficiency, which was unimaginable a few years ago. Virtualization holds a lot of promise - cost savings, server consolidation, and more efficient operations that will lead to higher quality of service. While these benefits paint a Pollyannaish picture, there are essential considerations to address to make the concept of virtual data centers fulfill its promise.

Management of virtual servers presents a new set of challenges that data center managers must understand to continue to grow with virtualization technology.

Challenges Facing Data Centers Going Virtual

There are several areas of concern that threaten the hope of virtual infrastructure for data centers. Virtualization introduces a new operating system and management layer – putting operations teams in the dark. The velocity of change complicates configuration and change management, multiplying the risks that arise from configuration drift or accident. As virtual integration expands its role in the data center , performance can be impacted.   

Virtual Machine Proliferation is the Biggest Concern

The performance demand for the multitude of virtual machines can overburdened the host server , risking slowed performance and possible crash of some or all the virtual machines connected to the server. Overburdening servers with excessive deployment of virtual machines can make this a serious issue for administrators. It can lead to failure of data protection resources as well. The provisioning of new virtual servers on the fly can easily lead to a situation of wild server sprawl. Stacking too many applications on one host physical server can leave them competing for resources. And managing your virtual resources among your physical machines can get complicated, especially for IT staff traditionally accustomed to managing only one application per physical server.

Data center admininstrators have to manage the capacity of servers and performance impact from virtual machine deployments. Without keeping a close check on changing configurations produced by  virtual machine proliferation, when performance slows occurs, the added time required to locate the root cause will delay incident response.  An automated solution is required to keep track of this volume and pace of change. 

Impact to Performance

Virtualization applies a layer of abstraction, decoupling from the hardware. With applications abstracted from the physical server hosting the virtual machine (VM), IT organizations are challenged in determining the root cause when an application stops responding. Many do not have the visibility necessary to discover what or how their applications changed on the VMs. The administrators can measure and report symptoms, such as poor performance but cannot diagnose the cause. By applying automated capabilities for analysis of changes and differences, down to detailed configuration parameters, focusing on criticality and impact, data center administrators can  make comparisons of environments, to see what changed and accelerate incident investigation. 

Managing Detailed Configuration Parameters

Virtual infrastructure while simplifying the data center with consolidation, adds many more parameters. This can cause users to experience low application performance. This demands more intense management and operations approaches for preventing unscheduled outages. Traditional configuration management comprises identifying configurations of system, hardware and applications; managing configuration changes; and collecting configuration information as part of ongoing operations.

Changes that are formal and informal, authorized and  not authorized, as well as unplanned changes to consolidated servers could have an impact that can cause outages to many parts of the business that can't be remediated by simply falling back on a snapshot image. For virtualized environments, a configuration management process is essential to retain the benefits and cost savings of consolidation. 

Virtualization and Managing Ongoing Growth 

Virtualization calls for new configuration management schemes, that will optimize resources and provide actionable information. IT organizations need to address configuration management in their virtual environments to ensure that they continue to enjoy virtualization’s promised benefits. This not only means addressing physical server configuration management needs, but also implementing an overall configuration management approach, incorporating virtualization's new dynamics.  

With advancements in virtual server technology, we are seeing expansion of virtual environments. Data centers need to focus on proper configuration management and a good understanding of the impact changes on the environment. An automated approach to challenges of maintaining stability in the virtual sphere can contribute to dynamic growth, and lay the groundwork for the oncoming cloud platforms.

About the Author
Martin Perlin