If your network has between 1,000 and10,000 devices and computers, you have a midsized network. Your servers, connections and other resources suffer the same problems as larger networks, but your budget for keeping the network healthy is less than what large enterprises enjoy.
Stories by Barry Nance
While SQL Server 2008 was little more than a service-pack level upgrade, the 2012 version of Microsoft's database has a boatload of new features and delivers solid performance improvements.
Virtually all our testing took place across 512Kbps frame relay, T-1 and T-3 WAN links. The test bed network consisted of six Fast Ethernet subnet domains linked by Cisco routers. Our lab's 50 clients consisted of computing platforms that included Windows 2000/2003/XP/Vista/Win7, Macintosh 10.x and Red Hat Linux (both server and workstation editions).
A network that measures downtime in millions of dollars per minute (or per second!) needs a serious, enterprise-level network management tool. Nothing less will do.
We evaluated each product in several different areas: Discovery and enumeration of devices and computers, support for a variety of device manufacturers and device types, global directory integration, graphical depiction of the network, monitoring of network node status (availability), performance and health, alerts and notifications when network problems occur, automated corrective actions, maintenance of trouble tickets (or integration with a help desk tool), support for virtualized environments and the production of useful, informative reports.
LANDesk Server Manager focuses tightly on the heart of your network -- its servers.
Do your network's cables and connections give you headaches? If so, the handheld NetTool and LinkRunner testers from Fluke belong in your network-management toolbox.
If you like the idea of a monitoring tool that displays a meaningful, easy-to-understand-at-a-single-glance map of your network and watches for connectivity problems like a hawk, then Dartware's Intermapper is the tool for you.
So the salesman tells you that once you buy the software, his engineers will come on-site and handle the installation, as it's standard practice for all customers. At first blush, you think you're saving time, effort and money when you take a software vendor up on its offer to sow its product across your network.
Playing chess, predicting the weather and simulating nuclear explosions are highly CPU-intensive applications. IBM's latest Deep Blue chess master, for example, is a 32-node IBM RS/6000 SP high-performance computer, and each node has eight onboard CPUs. The 256 processors work together to calculate 60 billion chess moves within three minutes, the time limit for each player's turn.
Installing firewalls, traffic prioritization tools and bandwidth governors typically involves buying separate products and configuring each via a convoluted, product-specific procedure. Having adequate expertise always on hand to deal with your network' s disparate software tools entails lots of cross-training, and the sheer number of products makes technical support a many-headed Hydra.
In a perfect world, your software vendor's (or your own company's) application programmers would precisely and fully understand the relationships between an application's consumption of computing resources and the available network bandwidth, server CPU speed, server disk speed and client CPU speed. Their grasp of these relationships would be so perfect they could tell you, for a specific set of network components, computers and users working at a given time of day, the exact response time for a particular application transaction or other unit of work.
Without connectivity, you don't have a network. When a network goes down, troubleshooters want to quickly learn which device is failing so they can restore network connectivity, while folks on the business side of the house need to know which workgroups and business functions are affected by an outage.
How you present information can be as important as the information itself. When it comes to network utilization reports, you need to make sure they are clear and easily understood if you want to convince your company that it needs to make network changes. Your reports can make the difference between a network with sufficient bandwidth and one with clogged arteries and high blood pressure.
Not so long ago, network managers would assign and manage IP addresses by manually updating tab-delimited HOSTS text files of static addresses and then distribute the resulting files throughout the company. Network administrators would put copies of the files in the appropriate directories for each server and each client. Many of these same companies added to the paperwork hell by mandating the manual assignment of a locally administered address for each network adapter, which overrode the burned-in network adapter ID. The resulting workload was a millstone around network administrators ' necks. Major hiring efforts and acquisitions of other companies crushed the administrators, whose lives became a horror show overnight (and over weekends).