Centralized Management
Today’s data centers contain a wide range of devices including application servers, email servers, database servers, Web servers, and e-commerce servers. These servers may consist of Intelbased servers, RISC servers, and even legacy minicomputers and mainframes. These servers run many operating systems including Windows, UNIX, Linux, and others.
The data center also includes numerous communication devices, such as routers, gateways, VPNs, and fi rewalls. And, all of these components require power, which is often managed by UPS devices, surge protectors, and power strips.
Business computing needs will always grow, which in turn increases the complexity of systems the IT administrator is responsible for managing. As an administrator, you have to add new hardware and systems as business needs require without sacrifi cing your existing management platform; you have to maintain additional servers without creating additional support problems.
In any medium to large organization, it’s rare for all of the network service components to be in one location. Branch and remote office environments often require key network components to be dispersed to many locations, sometimes all over the world. This can present an access problem for IT administrators.
Budget constraints usually mean that branch locations have very little in the way of IT support. It is often up to the IT administrator at the central offi ce to troubleshoot and fi x problems with systems that may reside hundreds or even thousands of miles away.
To be most effective, IT managers need a centralized management solution that gives them full remote access to all of the components regardless of their location. And, centralized management should mean more than just controlling each component from one desk. You should be able to see a combined view of all your connected servers and devices on a single computer monitor, in a single interface.
Network data systems are made up of disparate components from many different vendors. A useful remote management solution must be able to handle the needs of several different operating systems. It will not only provide control for PC-based systems, but also serial devices and AC power solutions.
Underlying all of these requirements is the need for security. Your management platform should secure access to your devices by leveraging the directory system you already have in place, allowing you to use a single user name and password repository. You should be able to assign device-level rights based on a user’s name so that administrators have access to more devices than an entry-level technician. Auditing and logging of all activity is also important so that you can keep track of “who does what” with your network infrastructure and all the devices attached to it.
Most vendors have a way of remotely accessing and managing theirequipment, but each tool is different. As the number of devices that need to be monitored, managed, and maintained grows, finding the right management tool or utility can be a chore.
For example, the typical enterprise will have network switches in racks that are normally managed using a Telnet session—either over a physical serial connection or via IP. Other devices, such as firewall and security appliances, are monitored and managed over an IP network using Web-based or Telnet-based management or through a third-party SNMP monitoring platform.
Intelligent Platform Management Interface (IPMI) technology allows IT administrators to monitor and control mechanical elements of their servers’ hardware, such as temperature, fans, and power supplies. Add keyboard, video, and mouse (KVM) switches and other commonly used tools such as remote management software, and the issue of access and control becomes quite complex.
Multiple tools mean multiple interfaces, multiple systems to learn, and no single point of management. Because there is no integration between the tools and no single interface to work from, you will have to connect using a method specific to that piece of equipment on your management console. This means you have to maintain multiple software clients, IP address lists, physical serial connections, and possibly other special hardware in order to connect to a device.
With multiple tools, the administrator has to become an expert on each different management solution. Learning curves are further exaggerated because each piece of hardware is a stand-alone piece of equipment that shares little or nothing with the other systems in the enterprise.
Without a centralized management console, there is no centralized user access management between each system. Each connection type must maintain its own user and password list—no sharing of information. A good centralized management platform should provide the benefi ts of single sign-on, where a secured authentication provides the necessary authorization and tracking for all of the tools, interfaces, and utilities that a specifi c IT administrator needs.
By combining your miscellaneous management tools into a single integrated platform, you can not only centralize your access control to each device, but also provide a smoother, more streamlined interface.
An effective solution must have access to all of these tools and components even when the network, server hardware, or the installed operating system is having problems. That means out-ofband management options that provide redundant access, because your management platform has to work all of the time.
Console management appliances must provide out-of-band access via dial connectivity. The right console management solution should let the appliance act as a PPP server. And it’s important that the appliance supports authentication and access control on dial connections. An integrated management solution speeds up the diagnostic process. Gathering information about a problem and then fi xing that problem can be accomplished faster when the IT administrator can turn to a single console with a unifi ed interface. When systems are diagnosed, repaired, and brought back online quickly, users experience very little downtime. They are more productive, the company runs more effi ciently, and the organization is more successful.