Welcome to
Virtualization and Beyond
Contributed by Michael Thompson, Principal Product Market
Management, SolarWinds
An Insurance Policy for Virtualized Mission Critical Apps
If you listen to the big vendors like VMware and
Microsoft, you would think that virtualizing mission critical applications like
transaction processing, reservation or sales applications would be an easy
decision for companies given the improved hardware utilization and increased
flexibility that virtualization provides. In addition, impressive performance
and capacity improvements in both VMware vSphere and Microsoft Hyper-V have
alleviated many of the basic resource concerns about CPU, memory and storage
input/output per second (IOPS). In addition, high availability, failover and
clustering also can reduce failure rates. But still, IT and business leaders
are cautious about the change. One reason is that both IT and business leaders
still know that IT systems go down and when they do go down it can be
expensive.
http://www.evolven.com/blog/downtime-outages-and-failures-understanding-their-true-costs.html#sthash.D1VaTtcW.dpuf
The reason systems still go down is that the most
common problems are not hardware or virtual machine failures. According to
Gartner, "Through 2015, 80 percent of outages impacting mission-critical
services will be caused by people and process issues, and more than 50 percent
of those outages will be caused by change/configuration/release integration and
hand-off issues." The reality is small human errors can quickly
turn into expensive and embarrassing business problems. That's why it is
extremely difficult to completely protect these systems with simply increasing
capacity or implementing redundant systems.
As a result, just like any other critical risk
that is hard to predict, IT and business leaders are looking for an "insurance
policy" when the stakes of virtualizing critical applications can become
financial survival or the loss of a good reputation. While you probably can buy
a literal insurance policy for such catastrophic IT events, I'm sure it's
expensive and would probably be cashed in while the CIO is out looking for a
new job. Instead, there are more practical things that an IT group can do to
help eliminate or mitigate outages when virtualizing mission critical
applications. In particular, a combination of three different approaches can
significantly reduce the risk.
- Leverage Appropriate HA and Failover
Technology and Procedures - HA and Failover technologies can increase
cost, not only for the HA or failover software but also for redundant
hardware and software licenses for standby systems. However, cloud
solutions and many companies' reduced cost for standby systems can help
alleviate this. That said, while technology can provide much of the
solution, you can't forget the people and procedures portion of the
equation. A surprising number of companies that purchase or set up
failover systems never test them. Alternatively, you can use
virtualization technology to provide a ‘poor man's' HA system with regular
snapshots, backups and religious maintenance of a set of golden master
images.
- Implement Virtualization-Appropriate
Change Management - Old-school process management approaches that lock
down any changes except those approved by a 20-person change committee
that meets once a month can be impossible in the dynamic virtualization
environment built on the ability to create, destroy or modify things in
seconds. However, it is possible to take change management concepts and
apply them to changes in virtual environments, including implementing
roll-back plans, snapshots and approvals for high risk activities.
- Leverage a Management System with
End-to-End Visibility and Configuration Tracking - No matter how much
technology gets implemented, a critical line of defense still is the
ability of an operator to monitor and quickly troubleshoot problems when they
occur, hopefully before they even impact end users. While many tools
provide a siloed view of one domain (e.g. application, virtualization or
storage), when virtualizing mission critical applications it is especially
important to be able to see performance data and root cause information
across silos in context of the application being virtualized. Products
like SolarWinds
Virtualization Manager with integration that provides in-context views
of application
performance and storage capacity
can be critical in ensuring that resource contention of shared resources
or changes in any layer will impact business critical applications.
Additionally, the ability to track historical configuration changes and
associate them with event data can be critical for finding problems that
result from human error, not hardware or software failure.
As more companies look to virtualize mission
critical applications, IT teams will need to carefully evaluate both the
technologies and procedures they use in the process to make sure that the
systems are highly available. Given the potential risks for business, companies
will need to carefully incorporate HA and failover planning, change management
and a system for increasing visibility and configuration management to provide
an extra insurance policy for both the business and their careers.
##
About the Author
Michael Thompson is a principal for product marketing management at
SolarWinds. Prior to this role, he served as director of business strategy for
virtualization and storage. Michael has worked in the IT management industry
for more than 11 years, including leading product management teams and
portfolios in the storage and virtualization/cloud spaces for IBM. He holds a
master of business administration and a bachelor's degree in chemical
engineering.