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.
Enterprises are converging more and more voice data and video dependent applications on the network. Understanding end to end network performance and being able to map and determine the users Quality of Experience is important. Path determination, trending of network utilization and the ability to baseline and then run “what-if” scenarios to understand the impact of deploying a new application is highly desirable.
