SHARE



ABOUT THE AUTHOR


Terry Slattery
Terry Slattery, is a senior network engineer with decades of experience in the internetworking industry. Prior to joining Chesapeake NetCraftsmen as...
Read Full Bio >>
SHARE



Terry Slattery | April 27, 2016 |

 
   

Recognizing the Need for More Bandwidth

Recognizing the Need for More Bandwidth Do you know when a network link needs more bandwidth?

Do you know when a network link needs more bandwidth?

When do you have to add more bandwidth to a congested link? When it runs out of bandwidth is the obvious answer. But how do you know that a link is congested? If you look at most network management link utilization displays, they rarely show anywhere close to 100% link utilization.

First, the effect of polling every few minutes effectively averages the link utilization over the time between each poll. Second, most products average several samples to arrive at aggregate data used in longer-interval displays. Figure 1 shows one-minute samples while Figure 2 shows the same data plotted with 10-minute resolution. The peaks are significantly less in the 10-minute plot because of averaging.

portable

Figure 1: One-minute utilization plot

portable

Figure 2: Ten-minute utilization plot

The 95th Percentile level, shown below in Figure 3 as horizontal blue (Receive) and red (Transmit) lines, displays more useful data. When calculated over 24 hours, the 95th Percentile shows the minimum utilization of the busiest 72 minutes of the day. Since 72 minutes is just a bit over an hour, I like to think of it as the minimum utilization during the "busy hour." In Figure 3, the Transmit 95th Percentile utilization is 25% and because the reporting interval is 12 hours, it shows the minimum utilization for the busiest 36 minutes of the day. Regardless of the type of measurement, utilization is not a good indication of congestion.

portable

Figure 3: One-minute utilization with 95th Percentile lines

Fortunately, there is a simple way to identify link congestion. Look for interface output drops (sometimes called "discards"). These occur when the network device has no free buffers in the output interface queue. Because packets awaiting transmission occupy all the buffers, the device must drop the new packet. Since a few drops are part of normal network operation, use the network management system (NMS) "Top-N" reports to show the interfaces that have the highest drop counts.

A second check is that these drop counts are greater than 0.0001% of link bandwidth. This is the point at which packet loss causes TCP performance to degrade. The combination of the Top-N drops and drop counts greater than 0.0001% identifies interfaces that are candidates for bandwidth upgrades.

Traffic Analysis
Now we need to determine if other tools, such as quality-of-service (QoS) monitors, can handle the congestion. Is there any important or time-sensitive traffic to prioritize? Conversely, can any packets be dropped (i.e., much less important or less time-sensitive data)?

You need to analyze the traffic using a tool that can show which applications are running so that you can prioritize the important applications and de-prioritize the least important applications. At one end of the tool spectrum is Wireshark, a basic packet capture tool that can perform application analysis (refer to this Sharkfest '12 presentation, "Application Performance Analysis"). Other tools include Riverbed's SteelCentral AppResponse for doing application performance analysis, and any of several types of flow data analysis tools.

At one customer, we were tasked to investigate slow application performance at a remote site. Our traffic analysis showed that entertainment traffic from Pandora.com, Akamai Networks, and Limelight Networks accounted for more than 50% of the 40-Mbps capacity available (Pandora is a streaming music site while Akamai and Limelight are content providers that deliver things like movies.) We de-prioritized the entertainment traffic into a low-priority queue, and gave higher priority to voice calls and business applications. This worked well at this site.

What About Buffering?
But couldn't the excess packets simply be buffered and transmitted when the congestion subsides? Well, that depends. If the congestion is of very short duration, buffering may work. However, do not use large amounts of buffering to try to control and reduce drops on heavily utilized links, especially if the links have low latency.

TCP will retransmit any unacknowledged packets after twice the round trip time (2 * RTT) because it thinks the packets have been dropped. My rule of thumb is to not use more buffers than the bandwidth-delay product of the link, in bytes, divided by the typical packet size (BW * RTT / typical-packet-size). It is important to identify a good packet size for use in this calculation. Voice packets are about 220 bytes long while file transfers are the maximum packet size (normally 1,500 bytes, but may be up to 9,000 bytes if jumbo frames are allowed). A value of 500 bytes is normally a reasonable starting point.

A good example may help you understand. Let's assume a 1Gbps link (125MBps) between two nearby data centers, with a normal RTT of 2 milliseconds (0.002 seconds). Most of the applications involved file transfers, so the typical packet size was close to 1,500 bytes. Plugging into our equation:

    1Gbps * 2ms / 1,500B = 125MBps * 0.002s / 1,500B = 166 packets
The typical hardware egress queue is 40 to 64 packets, so the maximum number of buffers to configure should be about 200.

If you configure an oversubscribed interface with too many buffers, it confuses the TCP retransmit algorithm. When queues build, the TCP retransmit timer expires and another copy of transmitted data is sent. This effectively reduces the link's "goodput" because it sends two copies of the same packet. So too much buffering actually hurts more than it helps.

When to Add More Bandwidth
If some of the traffic in our example could be put into a low-priority queue, then this might be a suitable solution. However, what if there isn't any low-priority traffic? That's when it is necessary to add more bandwidth. But how much bandwidth? Doubling the available bandwidth is typically a good starting point. I've not seen any calculations that would allow us to calculate how much bandwidth. It may be possible to calculate it using the Mathis Equation.

We encountered this situation at one customer. The link was dropping a lot of packets and was running at high utilization levels during working hours (this was an example where the NMS average utilization plots did show useful data). None of the applications were low priority, so we couldn't use QoS to prioritize the traffic. Adding bandwidth was the only recourse. Fortunately, the customer had already started the process of getting a link upgrade. Based on drop counts, we were able to determine that, at a minimum, the company needed to double link capacity.

What's That Again?
Interface utilization plots from network management systems are often not very useful for identifying congested interfaces. Instead, use interface drop counts, sorted from highest value to lowest value. Begin working on the Top-10 or Top-20 interfaces, focusing on those interfaces that show drop rates greater than 0.0001% of the link capacity (remember to convert interface speed from bits/second to packets/second, so you'll need a rough idea of packet size).

Analyze the traffic on the links with the most drops to determine the application mix. Identify any applications that must have high priority and those applications that can be given a low priority (i.e., that can be dropped when congestion occurs). If low-priority applications are consuming a significant amount of bandwidth, use QoS to force those packets to be dropped when congestion occurs. Check the QoS queue drop counts to determine if additional buffering is needed, but never add so much buffering that it confuses TCP.

If all applications are important and you can't select some traffic to drop using QoS, then you need more bandwidth. In this case, doubling the existing link capacity is a good start.





COMMENTS




October 24, 2018

With disparate workplaces and ever-expanding volumes of information to manage, the challenges for collaborating effectively are only intensifying. Many critical applications are not integrated, and

October 10, 2018

Businesses are growing across international borders quicker than ever, but scaling operations to follow suit can be a harder, longer process.

This webinar focuses on scaling your next-generat

September 26, 2018

Join Kevin Kieller, Microsoft UC&C expert, along with Ribbon Communications and Polycom, for an update on Microsoft Ignite, and a focus on critical things you need to know about your voice deployme

March 12, 2018
An effective E-911 implementation doesn't just happen; it takes a solid strategy. Tune in for tips from IT expert Irwin Lazar, of Nemertes Research.
March 9, 2018
IT consultant Steve Leaden lays out the whys and how-tos of getting the green light for your convergence strategy.
March 7, 2018
In advance of his speech tech tutorial at EC18, communications analyst Jon Arnold explores what voice means in a post-PBX world.
February 28, 2018
Voice engagement isn't about a simple phone call any longer, but rather a conversational experience that crosses from one channel to the next, as Daniel Hong, a VP and research director with Forrester....
February 16, 2018
What trends and technologies should you be up on for your contact center? Sheila McGee-Smith, Contact Center & Customer Experience track chair for Enterprise Connect 2018, gives us the lowdown.
February 9, 2018
Melanie Turek, VP of connected work research at Frost & Sullivan, walks us through key components -- and sticking points -- of customer-oriented digital transformation projects.
February 2, 2018
UC consultant Marty Parker has crunched lots of numbers evaluating UC options; tune in for what he's learned and tips for your own analysis.
January 26, 2018
Don't miss out on the fun! Organizer Alan Quayle shares details of his pre-Enterprise Connect hackathon, TADHack-mini '18, showcasing programmable communications.
December 20, 2017
Kevin Kieller, partner with enableUC, provides advice on how to move forward with your Skype for Business and Teams deployments.
December 20, 2017
Zeus Kerravala, principal analyst with ZK Research, shares his perspective on artificial intelligence and the future of team collaboration.
December 20, 2017
Delanda Coleman, Microsoft senior marketing manager, explains the Teams vision and shares use case examples.
November 30, 2017
With a ruling on the FCC's proposed order to dismantle the Open Internet Order expected this month, communications technology attorney Martha Buyer walks us through what's at stake.
October 23, 2017
Wondering which Office 365 collaboration tool to use when? Get quick pointers from CBT Nuggets instructor Simona Millham.
September 22, 2017
In this podcast, we explore the future of work with Robert Brown, AVP of the Cognizant Center for the Future of Work, who helps us answer the question, "What do we do when machines do everything?"
September 8, 2017
Greg Collins, a technology analyst and strategist with Exact Ventures, delivers a status report on 5G implementation plans and tells enterprises why they shouldn't wait to move ahead on potential use ....
August 25, 2017
Find out what business considerations are driving the SIP trunking market today, and learn a bit about how satisfied enterprises are with their providers. We talk with John Malone, president of The Ea....
August 16, 2017
World Vision U.S. is finding lots of goodness in RingCentral's cloud communications service, but as Randy Boyd, infrastructure architect at the global humanitarian nonprofit, tells us, he and his team....
August 11, 2017
Alicia Gee, director of unified communications at Sutter Physician Services, oversees the technical team supporting a 1,000-agent contact center running on Genesys PureConnect. She catches us up on th....
August 4, 2017
Andrew Prokop, communications evangelist with Arrow Systems Integration, has lately been working on integrating enterprise communications into Internet of Things ecosystems. He shares examples and off....
July 27, 2017
Industry watcher Elka Popova, a Frost & Sullivan program director, shares her perspective on this acquisition, discussing Mitel's market positioning, why the move makes sense, and more.
July 14, 2017
Lantre Barr, founder and CEO of Blacc Spot Media, urges any enterprise that's been on the fence about integrating real-time communications into business workflows to jump off and get started. Tune and....
June 28, 2017
Communications expert Tsahi Levent-Levi, author of the popular BlogGeek.me blog, keeps a running tally and comprehensive overview of communications platform-as-a-service offerings in his "Choosing a W....
June 9, 2017
If you think telecom expense management applies to nothing more than business phone lines, think again. Hyoun Park, founder and principal investigator with technology advisory Amalgam Insights, tells ....
June 2, 2017
Enterprises strategizing on mobility today, including for internal collaboration, don't have the luxury of learning as they go. Tony Rizzo, enterprise mobility specialist with Blue Hill Research, expl....
May 24, 2017
Mark Winther, head of IDC's global telecom consulting practice, gives us his take on how CPaaS providers evolve beyond the basic building blocks and address maturing enterprise needs.
May 18, 2017
Diane Myers, senior research director at IHS Markit, walks us through her 2017 UC-as-a-service report... and shares what might be to come in 2018.
April 28, 2017
Change isn't easy, but it is necessary. Tune in for advice and perspective from Zeus Kerravala, co-author of a "Digital Transformation for Dummies" special edition.
April 20, 2017
Robin Gareiss, president of Nemertes Research, shares insight gleaned from the firm's 12th annual UCC Total Cost of Operations study.
March 23, 2017
Tim Banting, of Current Analysis, gives us a peek into what the next three years will bring in advance of his Enterprise Connect session exploring the question: Will there be a new model for enterpris....
March 15, 2017
Andrew Prokop, communications evangelist with Arrow Systems Integration, discusses the evolving role of the all-important session border controller.
March 9, 2017
Organizer Alan Quayle gives us the lowdown on programmable communications and all you need to know about participating in this pre-Enterprise Connect hackathon.
March 3, 2017
From protecting against new vulnerabilities to keeping security assessments up to date, security consultant Mark Collier shares tips on how best to protect your UC systems.
February 24, 2017
UC analyst Blair Pleasant sorts through the myriad cloud architectural models underlying UCaaS and CCaaS offerings, and explains why knowing the differences matter.
February 17, 2017
From the most basics of basics to the hidden gotchas, UC consultant Melissa Swartz helps demystify the complex world of SIP trunking.
February 7, 2017
UC&C consultant Kevin Kieller, a partner at enableUC, shares pointers for making the right architectural choices for your Skype for Business deployment.
February 1, 2017
Elka Popova, a Frost & Sullivan program director, shares a status report on the UCaaS market today and offers her perspective on what large enterprises need before committing to UC in the cloud.
January 26, 2017
Andrew Davis, co-founder of Wainhouse Research and chair of the Video track at Enterprise Connect 2017, sorts through the myriad cloud video service options and shares how to tell if your choice is en....
January 23, 2017
Sheila McGee-Smith, Contact Center/Customer Experience track chair for Enterprise Connect 2017, tells us what we need to know about the role cloud software is playing in contact centers today.