Inside The Stack

Mainframe Product

Inside the stack

More Information

You may want to also see our Problem Finders or z/OS monitoring products. Use the site navigation bar above or click on the product name below.

For the Problem Finder products, please go to https://www.ipproblemfinders.com.

Inside the Stack is an expert system, providing case studies and structured knowledge integrated with live data from your z/OS system so you can solve complex performance problems quickly. Inside the Stack provides a complete TCP/IP monitor. Is your boss asking you how much FTP traffic you have? Is TCP/IP a black box to you? With Inside the Stack, you can truly view what is going on inside the TCP/IP stack!

You can use Inside the Stack every day to diagnose and resolve problems on your z/OS TCP/IP network. If you are responsible for managing a mainframe-based network, don’t try to do it without Inside the Stack!

  • Take a look at the slides below.
  • You will see tuning we did using Inside the Stack
  • We were able to save approximately 30% of the CPU for TCP/IP

Mainframe

We welcome your questions and queries. Please see our Contact Us page for complete contact information.

Inside the Stack

  • Just as the arteries of the heart can become clogged, so can the
    network.
  • Inside the Stack points you to the area of TCP/IP which needs tuning.
    Then, you may use our other products to find the exact problems.
  • Problems such as: low throughput, poor response time, network
    overhead. excessive mainframe CPU usage.

After Tune-Up

  • After a Health Check and tuning efforts lasting 2-3 weeks, the listener and
    UP errors for both systems have been completely eliminated.
  • The IMP errors for both systems are nearly eliminated
  • The TCP errors have been cut to 1/4 to 1/3 of what they used to be

A Recent Health Check

  • Data from a recent Network Health Check reveal TCP, UP, ICMP, and
    listener errors for both systems.
  • Over 2,000 errors per 3-minute interval.
  • With tuning these numbers fall significantly.
  • Errors contribute to TCP/IP RB usage.

CPU Usage Before Changes

  • TCP/IP is the 2nd highest
    user of SRBs.
  • Very repeatable.

CPU Usage After Changes

  •  TCP/IP is now 4th
    highest instead of 2nd.
  • Only 2 or 3 weeks of
    tuning effort.

20% Create 80% of Problems

  • Often, a few users or devices are
    responsible for the bulk of the
    problems on your network.
  • We have found one user with an
    improper script responsible for
    40% of the connections on one
    application
  • We have found one
    communications link causing
    millions of retransmissions. Over
    75% of the retransmissions on this
    company’s network were due to
    one device significantly.