The principles and benefits of application server virtualization are well understood. At the very least, the consolidation of physical resources can drive significant capital and operational expenditure savings. Although the scope of virtualization programs vary between different organizations, most businesses are trying to improve the efficiency, agility and the resilience of their data centers - in addition to driving for greater cost savings.

In order to truly profit from the benefits derived from the virtual data center, a complete architecture redesign of all layers of the data center is needed. The first layer is virtualization and consolidation of the server and storage infrastructure. According to recent surveys, most organizations have deployed server virtualization but have not yet virtualized 100% of their environment, creating hybrid environments of dedicated and virtualized servers.

The second layer is the virtualization and consolidation of network and infrastructure hardware and the transition from a siloed architecture to a flatter, less tiered virtualized network design. Infrastructure virtualization and consolidation has a major impact on the role of the Application Delivery Controller. In the virtualized data center, new services can deployed quickly and configuration changes addressed instantly. To avoid negating the benefits of virtual server technology, ADC resources should also be able  to be provisioned instantaneously; without the need for protracted physical equipment deployment and configuration. Creating a virtualized Application Delivery infrastructure layer allows alignment of the Application Delivery services with the frequent changes in the virtualized data center, and also provides each application with an Application Delivery service matching its SLA and performance predictability needs.

The shift from siloed and sprawled data centers designs to consolidated, flatter architectures has placed new requirements on Application Delivery Controllers. Radware has addressed the need with its new Virtualized Application Delivery Infrastructure strategy; enabling the consolidation and virtualization of Application Delivery services as an integral part of the virtual data center architecture, its orchestration and provisioning systems. It’s Application Delivery Controller product line has been transformed so that virtual ADC instances (vADCs) now can run on top of specialized or general purpose computing resources. Three different form factors are supported:

i. Dedicated ADC – a dedicated physical ADC device running a single vADC

ii. ADC-VX – multiple vADCs running on top of a specialized ADC hypervisor and dedicated ADC hardware

iii. Soft ADC - vADC instances running as virtual appliances on general hardware

ADCs play a special role in the data center; they ensure the performance and quality of experience of applications and, as such, must demonstrate highly predictable performance, have its own dedicated and isolated resources. Radware’s ADC-VX product delivers the best of both the virtual and physical world’s; the financial and architectural benefits of platform consolidation and virtualization but with the predictable performance and resiliency of physical units.

As with server consolidation, ADC consolidation can achieve significant cost savings. Users of virtualized ADC-VX technology can expect to see total cost of ownerships savings of over 30% compared to physical ADC implementations.

  Assuming that the best decision-making begins with the end in mind, an IT organization should analyze its capabilities, tools as well as the needs and goals of the organization upfront.  For example, what are the requirements, and metrics, and other factors should they get clear on before evaluating monitoring solutions?

  Network system and solution monitoring has typically been an afterthought, and this is because monitoring, analysis and troubleshooting requires more of an operational skillset and mindset, which needs to be more proactive. 

  Looking for issues before they are business or end-user impacting productivity requires resources that most organizations either have not invested in, or are less willing to fund. 

  Organizations should determine what their key systems are, where their level of risk exists from system failure, and interview end-users, business unit leaders, and system managers to identify and analyze the organizations monitoring requirements.

Factors might lead IT to conclude that contracting with a monitoring service provider is a better solution than purchasing monitoring software.  Historically, as a company grows, reliance on technology also increases, but should the company shrink its reliance, monitoring needs doesn’t lessen but may actually increase further.

  Historically management and monitoring systems have been an afterthought for organizations, and since they typically require multiple full time employees to manage and maintain effectively, along with the price of acquisition, maintenance and support of the monitoring system itself; monitoring is one of the most cost effective information technology requirements to outsource.  More and more organizations that have had large extensive monitoring systems, which as they get older or as budgetary cuts and staff reduction have impacted the effectiveness, also move from internal monitored to an external monitoring service 

  When IT has decided that it makes more sense to go with a monitoring service provider, what questions should they ask, and what aspects about a potential vendor should they look at, to determine if the match is going to be a good fit? Organizations should primarily focus on determining what their requirements today, and not only ensure that the provider has the capability to manage each component today, but ask about other systems for solutions or services that the organization is either contemplating deployment or is interested in as well. 

  It is important to make sure that the potential service provider is also forward thinking, and that their direction matches that of your organization.  Most monitoring opportunities turn into managed services opportunities for like, or new technologies.  Determining upfront if your monitoring service provider could meet your needs should also include this flexibility.  More importantly, integration and flexibility of the monitoring system to integrate into your environment is important. 

  Being able to tie into your existing authentication services may be nice to have, but integration with your ticketing or CRM utility may be more critical.  Most of the newer cloud based monitoring service focus primarily on the Windows Servers and desktops and their applications.  Enterprises should not assume that the monitoring service provider has the basic capabilities of importing SNMP MIBs to monitor and control any devices. 

  Other technical issues of concern go into IPv6 and SNMPv3 Support and even Syslog data, the monitoring service should be able deploy onsite data collectors taking in Netflow, JFlow or IPFix type application performance data that can be collected at the network level.  This is important to determine and identify where application performance issues occur, and to monitor user experience including determination of MOS scores as routinely understood now as part of IP Telephony and Unified Communications.