CMP's TechWeb Click Here to Visit Mentor Labs


Data Communications
Search Data Communications

 Browse By...
 Technology/Topic
 Vendor
 Issue

 Visitors Center

 FAQs
 Contact the Editors
 Registration
 Subscriptions

 Content
 TECH TUTORIALS
 ATM
 Carrier-Class Gear
 Internet/Intranets
 IP Tools/Issues
 Network Software
 Net Management/SLAs
 QOS
 Remote Access
 Routers/Switches
 Security
 Servers/Peripherals
 Services/Outsourcing
 Workgroup Networks

 NEW PUBLIC NETWORK

 LAB TEST CENTER

 GLOBAL NETWORKS

 PRODUCT LEADERS

 OPINIONS/COLUMN
 Viewpoint
 Lippis on Internetworking
 Sound Byte

 Sales & Advertising
 Editorial Calendar
 Reach the Sales Reps

 Marketing Services
 T99 Media Services
 Marketing/Subscriptions
 Industry Front & Center
 Reader Service
 Card Deck

 Real-World Seminars
 Schedule
 Sponsorships

 Custom Publishing
 Vendor Strategies
 Sponsorships

Click here to visist CMPnet Radio.

TechWeb Sites
 Byte.com
 Data Communications
 File Mine
 InformationWeek
 InternetWeek
 Network Computing
 Planet IT
 TechShopper
 TechWeb News
 Tele.com
 Winmag.com

September 7, 1999


By Mary Jander

Product Leader: Network Management

Delivering the Right Amount of Data on Distributed Nets

Never enough, or way too much. Those tend to be the choices when it comes to collecting management data on distributed IP nets. Event correlation apps skimp on the stats, while leading management consoles tend to go into information overload.

System Management Arts Inc. (Smarts, White Plains, N.Y.) says its Incharge 3.0 event correlator delivers just the right amount of data. Its ability to view network events extends to servers and applications, as well as routers and switches. Incharge 3.0 not only reports the root causes of network problems, but also indicates their severity and highlights the devices and apps most likely to be affected by the trouble. And it doesn’t require customers to spend $15,000 on an SNMP console (although it can use SNMP data if it’s already available). Instead, it automatically discovers network elements on its own.


MORE INFO
Product Summary



But for all of that, Incharge 3.0 can’t delve into specific causes of app-related problems, such as memory shortages or CPU malfunctions. Nor is it integrated with Incharge’s frame relay analysis software. A GUI is MIA, and customization tools are costly.

The Discovery Phase

Incharge consists of software that runs under Windows NT or Unix. It performs automatic discovery of devices and applications in the network using a range of techniques, from pinging remote machines to deploying command-line interfaces for specific devices.

Incharge typically gathers application data by checking which TCP/IP port numbers are active, or by accepting input on network topology from a management console, such as Openview from Hewlett-Packard Co. (HP, Palo Alto, Calif.). Net architects also can prepare a file of specific IP address as input. This information can then be used as the basis for ongoing monitoring and event correlation.

Once the network is discovered, information about network elements is transferred into an integral topology database. Smarts has rewritten this database from the previous version of Incharge so that it’s possible to view devices in the switched (Layer 2) and routed (Layer 3) portions of the network, and to gauge the performance of apps that rely on these devices.

Next, Incharge makes use of a range of protocols (including SNMP and FTP) to check the availability and condition of devices, servers, and applications. If a router or switch fails and generates an SNMP alert, for instance, Incharge immediately picks up this information—along with all of the other events associated with the outage of that device.


On Top
of Trouble


This event information is matched up with the database input and run through the package’s diagnostic algorithms. Only the most significant events actually appear on the administrator’s screen. And among those, the most likely root cause of the problem is identified (see the figure). The result? A way to weed out extraneous alerts while getting to the heart of network trouble quickly—and speeding up recovery from downtime.

Combo Kudos

By combining Layer 2 and Layer 3 diagnostics with the ability to capture information on application traffic, Smarts is ahead of the competition. Neither Eye of the Storm from Prosum Ltd. (London) nor Openriver from Riversoft Ltd. (London), for example, track app and server status.

And early users are pleased with that mix of features. "This release takes Incharge to another level," says Mark Leighty, an internetworking consultant with integrator RPM Consulting Inc. (Columbia, Md.) who has installed the software in a pilot project. "I haven’t seen this combination before."

Incharge also makes it easy to group alarm reports by application or department. And net managers can choose the problems they wish to monitor, or selectively filter events according to various criteria. For instance, they can opt to see only the most severe alarms or only those associated with particular portions of the network for which they’re responsible.

But there are some drawbacks prospective users should keep in mind. Incharge isn’t integrated with Smarts’ Frame Relay Manager, an application that has gained considerable popularity. Incharge also is short a GUI. Smarts furnishes screens of text-based, color-coded messages instead of the map-based illustrations of alerts many net architects prefer.

Incharge 3.0 runs under Windows NT, Solaris, AIX, and HP-UX; pricing starts at $25,000.



Home Contact Editors Lab Tests Registration Tech Tutorials
Buyer's Guide Global Networks Opinion / Columns FAQs Subscriptions

CMPnet