What is “Managed Services”
This article was first published in the March 2013 Issue of Technology First Magaine.
Barry D. Hassler
There are lots of buzz-words floating around the entire technology industry, and the computer support services side of technology is no exception. We have great terms like “IT”, “cloud”, “BYOD”, and “Managed Support”. For those in the IT industry, these terms are at least familiar, but I find even that familiarity will often come with a total misunderstanding of what the terms means, how it’s used, when it’s appropriate, or what the pitfalls are.
I’ll concentrate on the “Managed Support” one for now. Much of the information I provide here is based not only on industry information, but on HCST’s extensive experienced as a recognized Managed Services Provider for over 5 years, and our even longer experience in providing proactive support for our wide range of clientèle throughout the greater Miami Valley.
What is Managed Services?
There are two primary ways in which any company can support their computer systems (“IT” – Information Technology). You either handle problems as they occur in a reactive (it’s broken, let’s fix it) fashion, or take a more preventive approach to prevent problems or detect issues before they become problems. The idea behind the latter is that the preventive approach should increase reliability and improve overall efficiency and improve operations. The latter is commonly referred to as “Managed Services”. The more reliable the overall IT infrastructure (including networks, servers, workstations, printers, etc) is, the more productive the users of that resource can be. On the other hand, in the reactive approach, productivity is already being lost even before someone cries “help!”.
In a broad sense, “Managed Services” can be applied to not only just computer support, but other related areas such as applications, Internet connectivity, printers, backups, security, telecommunications and other IT related technologies.
The Managed Services concept includes extensive monitoring, automated “self-healing” procedures, escalated notifications as well as manual intervention. All these predicated around detecting issues before they become problems.
A Managed Services Provider (MSP) will generally use a combination of systems and technologies to implement their Managed Services offerings. These will include: network monitoring systems employing a variety of monitoring protocols such as the Simple Network Monitoring Protocol (SNMP) Windows Management Interface (WMI), log analysis, and specialized scripts; automated ticketing systems which track issues, responses, and resolutions (and generally billing); a library of industry “best practices” which provide a knowledge base of correct implementation, trouble shooting and problem resolutions. Lest anyone think Managed Services can be a totally automated or scripted service, we find highly qualified and experienced technicians to be the most important part of the Managed Service Provider’s arsenal.
Managed Service Providers will use commercial or open-source network monitoring systems which are able to monitor a wide variety of data points across a broad range of equipment. While I am a ardent proponent of open-source technologies, the breadth of capabilities and support of a commercial network monitoring system is essential to the successful MSP practice.
Monitoring is accomplished primarily using SNMP or WMI from a centralized network monitoring system, or distributed “probes” or agents deployed to a customer’s network. In our case, HCST deploys “probe” software to each client’s network, and monitoring “agents” on monitored servers and workstations. The probes communicate with other agents on the client’s network to accomplish the wide variety of monitoring tasks, and coordinate communication back to the centralized network monitoring system. Probes and Agents perform direct monitoring of such items as cpu, memory and disk utilization, process status (determining if processes are running or not), log analysis (detecting particular messages in logs), as well as specific monitoring points in windows services such as Exchange, SQL Server, IIS and other applications.
The frequency of monitoring is dependent on the type of system, importance, or type of service. For instance, connectivity to a server may be be monitored every few minutes (or in some cases, seconds), while monitoring the status of a daily backup would be performed only daily. Disk utilization might be monitored on an hourly basis.
Automated “Self Healing” procedures
In HCST’s case, implementing “self-healing” procedures is a first step in the resolution of common problems. For instance, if monitoring detects a critical service has stopped running, and initial self-healing procedure will attempt to restart the process automatically before proceeding with further notifications to support staff. In almost all cases, these self-healing procedures can be successful in quickly resolving issues.
The third critical component to the Managed Services is automated notifications with appropriate escalations. While detection of issues is critical, unless a self-healing procedure can resolve the issue, human technicians must be notified in a timely fashion depending on the particular issue. The ability to repeat notifications until resolved ensures an on-call technician(s) do not “miss” notifications, and escalations ensure high-level staff or management are notified as well for certain types of events, or continuing issues. As in the case of monitoring frequencies, the frequency at which notifications are provided should be dependent on the particular issue.
A successful Managed Service Provider naturally needs a staff of qualified and experienced technicians who are well versed in the variety of technologies and components being used by their clients. MSP technician’s need to be able to resolve the issues detected by the variety of monitoring systems in a timely fashion.
In many cases, an MSP’s relationship with a client will be based on uptime, system or process availability or time-to-respond metrics which are used to determine the success of the Managed Services. The Managed Service Provider depends on in-house staff and appropriate out-sourced support personnel to ensure the quick and accurate resolution of any issues.
Managed Services aren’t always the right solution for every business. Each business needs to evaluate their requirements in conjunction with their IT Services provider (in-house, or outsourced) to determine the impact of various outages of IT resources, and the resultant down-time of their staff or services.
When any outage of network servers, workstations or Internet could result in staff being unable to perform their duties, or inability to provide services to clients, Managed Services are a potential solution to increase reliability of those resources, or decrease the downtime associated with outages.
Some simple examples of situations which could easily be prevented by use of the Managed Services described here may help clarify the successes associated with these services.
Microsoft Exchange databases can be easily corrupted if the file system containing the exchange data stores fill up. If these applicable partitions are not monitored, filling them up can result in extensive recovery efforts which may entail several hours of email downtime. If your business depends heavily on email services, extensive monitoring of these critical services will ensure high availability of those services.
HCST can attest to both sides of this situation – with a number of situations where monitoring could have prevented hours (or even a day) of email outages. Likewise, with our monitoring capabilities, HCST is able to detect uncontrolled growth of exchange datastores, or failed backups which do not roll Exchange log files.
Monitoring the status of backups can ensure an organization has the ability to recover from lost data. In the above situation with Microsoft Exchange, imagine the impact of a Microsoft Exchange datastore filling up, and not having a successful backup for several days. Loss of days worth of email could be disastrous for an organization.
Just changing tapes, USB drives, or assuming daily “cloud” backups complete successfully is not sufficient. Unless automated solutions are put in place (possibly as part of a Managed Services package), manual review on a daily basis is necessary to ensure this critical business function.
Loss of services due to “crashed” services
A failed service on a Windows service can in some cases produce a difficult to diagnose loss of service. The use of self-healing procedures can result in a quick and painless resolution to this type of event.
A case in point is a client with a licensing service for a third-party application which is not as stable as it should be, resulting in the service occasionally shutting down without warning. By monitoring process status, HCST is able to quickly restart the service without human interaction in a matter of minutes. Without this capability, the service crash would cause users to be unable to use the service until a technician can respond and restart the service.
In addition to the above mentioned case studies, there are several other benefits clients can derive as the result of a Managed Services contract with a competent Managed Services Provider. These benefits include the somewhat obvious increased efficiency as the result of less downtime and higher availability of services as well as reduced downtime and risk as the result of the extensive monitoring and proactive responses.
Other benefits that may not be quite as obvious are capacity planning information as the result of detail statistics accumulated by the Managed Services Provider, up-to-date patch management which can be automated, or controlled better through a Managed Services platform and a better understanding of the IT infrastructure. Managed Services platforms keep records of servers, workstations, printers, network devices and any other critical devices. The platform can provide detailed inventory information, including installed applications and versions.
Another huge benefit to a Managed Services approach to IT support is the predictability of costs, where the costs of monitoring and fixed services is generally a fixed monthly charge, instead of a highly variable cost of more reactive-based services.
Most Managed Service provides (HCST included) prioritize responses to Managed Services clients over reactive clients, and further, are able to use remote control utilities more readily to respond to issues remotely instead of waiting for a technician to arrive on-site.
The use of Managed Services to support a critical IT infrastructure can be highly successful for appropriate organizations in terms of increasing reliability, decreasing downtime, and providing a more predicable cost for IT services.
While not necessarily applicable to ALL organizations, HCST reviews each clients needs and requirements in detail with them to provide the appropriate level of services in a Managed Services environment.