Technical Brief: Capacity Planning | NETSCOUT

Technical Brief: Capacity Planning

Business Challenge

In any distributed organization, the cost of connectivity is always difficult to quantify. Traffic on the network tends to expand to fill the available bandwidth, whether or not that traffic is crucial to the business. Usage of business-critical applications is also difficult to measure. It may be within expected parameters for 90% of the time, but no business wants to experience degraded performance for the remaining 10%.

End-user-experience performance requirements put constant pressure on IT to increase bandwidth in order to provide an effective service to the business, while cost management requires that this is limited, or even reduced. So, how can an IT organization manage these conflicting pressures?

Technical Challenges
  • To identify a link that requires additional bandwidth, you must demonstrate that it is experiencing congestion for a significant amount of time.
  • It can be costly, time-consuming (sometimes, depending on the tools used, even impossible) to correctly represent congestion over long time periods.
  • Congestion must be caused by legitimate business usage, not recreational or rogue usage (“other” traffic): bandwidth spending is the final resort, removing “other” traffic from the network comes as a much earlier option.
Common Approaches
  • Long-range views of average utilization: this will show a long-term trend of utilization, but the long-term view will average out those spikes of high utilization, thus hiding the problem.
  • Peak utilization, e.g. showing the busiest minute for each day in a month: This shows what days had a busy minute, but doesn’t give insight into the amount of time during which time a link is congested.
  • Traffic totals: easy to show all links on a single view, showing the links with most traffic and even periodic trends to show month-by-month usage. However, it does not give any indication of congestion except in extreme cases.

The TruView Solution – Simple, Accurate, Effective

Use the Capacity Overview to identify which links have been over-utilized, and for how much time. Those links that were over 80% utilized for more than a threshold timeframe are highlighted with a red icon.

  • Using a Burst-style view, TruView displays the amount of time for which utilization is within each utilization band, from <30%, 30–60%. 60–80% and >80%. This clearly and simply shows what links spent the most amount of time in a congested state (<80%).
  • Based on 1-minute measurements, detail of congestion is never lost for views of months, weeks or even a year.
  • “Follow the red” to see the most congested links.
  • Reverse the ordering to see those links that were never utilized over 30% and thus may be downgraded.
  • Include a Business Hour filter to examine only daytime usage, eliminating interference from nightly backup and update jobs.
  • For those links which are not over-utilized, use >60% levels to help build long-range budget estimates.

View the utilization levels over time on the Capacity Detail page, and identify who and what made the links busy with Application and Host breakdowns.

  • Use a single-link view to easily determine if congestion is caused by valid business usage or unauthorized / recreational use.
  • SNMP and NetFlow data are combined to show when a link is busy, how busy it is, but also why it’s busy.
  • Burst detail broken out to a time trend to show if usage is stable, erratic or growing.
  • Simple-to-understand view, can be used as part of a business case for increased bandwidth spending.