GTWP TAP Vs SPAN21
GTWP TAP Vs SPAN21
GTWP TAP Vs SPAN21
WHITEPAPER
SPAN
Best Practice
Guide to Improving
Network Visibility
HOW TO NAVIGATE
NETWORK VISUALIZATION
This whitepaper is an in-depth look into network
visualization access.
A network TAP provides strategic, persistent monitoring capabilities. Installing a TAP during deployment means
you have a permanent method of access to network traffic.
• Ensure 100% full duplex copies of network traffic without altering the data.
• Support 10M, 100M, 1G, 10G, 40G, 100G, and 400G.
• Are scalable and can either provide a single copy, multiple copies (regeneration), or consolidate traffic
(aggregation) to maximize the production of your monitoring tools.
EMA recommends that enterprises use TAPs as much as possible in the access layer to
avoid network performance impacts and assure packet fidelity.”
Many switches have a limit on the number of SPAN monitoring ports that you can configure. This limit is often a
maximum of two monitoring ports per switch.
Port mirroring best practices vary by switch vendor, as many architectures use non blocking methods that drop
overages if you overrun a port mirror, depending on the switch you use, there can be an adverse effect on traffic
or switch performance.1
• SPAN are programmed ports on a switch, that provide access to packets for monitoring.
• SPAN sessions do not interfere with the normal operation of the switch.
• Low priority processing — the switch will drop SPAN packets if heavily utilized or oversubscribed.
SPANs can add overhead on a network device, and that SPAN port will often drop
mirrored packets if the device gets too busy. Therefore, TAPs are a better option.”
1. We have entered a much higher utilization environment with many times more frames in the network.
3. We have entered into the era of data security, deep packet capture, legal and policy compliance, network
auditing and Lawful Intercept approved by Commission on Accreditation for Law Enforcement Agencies
(CALEA) which requires that we must monitor all of the data and not just “sample” the data, with the exception
of very focused monitoring technologies (i.e. application performance monitoring).
These demands will continue to grow since we have become a very digitally focused society and are all
connected via the Internet of Things (IoT). With the heavy use of VoIP and digital video we have revenue
generating data that is connection oriented and sensitive to bandwidth, loss and delay.
The older methods need reviewing and the added complexity requires that we change some old habits to
allow for real 100% Full Duplex real time access to the critical data. Being able to provide real access is not
only important for Data Compliance Audits and Lawful Intercept events, it is the law.
Diagram 3: Cisco reference diagram for their Catalyst switch SPAN port.
Knowing that the SPAN port arbitrarily drops traffic under specific load conditions, what strategy should users
adopt so as not to miss frames? According to Cisco,
the best strategy is to make decisions based on the traffic levels of the configuration and
when in doubt to use the SPAN port only for relatively low-throughput situations.”2
Also, consider that a switch’s SPAN access is not fault tolerant and can be a major fault or failure point for your
monitoring and management vision. A network TAP is not a failure point.
This does not cause any switch performance degradation, disruption or traffic flow
on the source ports. The only affected port is the destination port, and it drops
packets on a first in first out (FIFO) basis once the egress buffer limit is exceeded.
• Monitoring tools may miss packets due to SPAN port oversubscription.
• SPAN will not pass corrupt packets or errors (bad packets), these are dropped.
You can SPAN packets in or out of a switch port, but typically most applications
require a copy of both sides. SPANs are known to result in trace files with
duplicated packets when the SPAN port is set up to capture both ingress and
egress traffic flows. This common problem when both the ingress and egress
ports are spanned, end up sending duplicate packets to the monitoring tool, which
becomes a whack-a-mole type headache.
• SPAN can change the timing of the frame interactions, altering response times
• The timestamps are can read different but the packet contents are the same
• Can duplicate packets if multiple VLANs are used
This has all changed with 1, 10, 40 and 100 Gigabit technologies starting with the
fact that maximum bandwidth is now twice the base bandwidth – so a Full Duplex
(FDX) Gigabit link is now 2 Gigabits of data and a 10 Gigabit FDX link is now 20
Gigabits of potential data flows.
No switch or router can handle replicating/mirroring all of that data, plus handling
its primary job of switching and routing. It is difficult if not impossible to pass
all frames (good and bad ones), including FDX traffic at full time rate with the
interframe gap, in real time at non-blocking speeds. All of this, on say 16 ports, is a
whole lot of data to go through one port. Furthermore, to this FDX need we must
also consider the VLAN complexity and finding the origin of a problem once the
frames have been analyzed and a problem detected.
These monitoring requirements utilize a small amount of bandwidth and grooming does not affect the quality
of the reports and statistics. The reason for their success is that they keep within the parameters and capability
of the SPAN ports, and they do not need every frame for their successful reporting and analysis. In other words,
SPAN port is a usable technology if used correctly and the companies that use mirroring or SPAN are using it in a
well-managed and tested methodology.
SPAN could also be used in a remote location that doesn’t justify a permanent deployment, offering temporary
access for troubleshooting. After all, SPAN ports were not intended for long-term use.
USE CASE
Network TAPs can enhance SPAN deployment
Yes, network TAPs provide 100% visibility to your network monitoring and security tools, but as SPAN is still being
used, there are many TAP use cases that can enhance your current SPAN deployment. There are many situations
when there are not enough SPAN/mirrored ports available on a router or switch to allow access to all of the
monitoring tools that need to see the traffic of the link. Introducing a Regeneration/SPAN Mode TAP provides a way
to distribute a link’s traffic to up to multiple network tools.
SPAN Regeneration: Garland’s network TAPs have SPAN or regeneration mode, which allows you to take one
SPAN link and copy the same traffic to multiple tools (1:3, 1:5).
SPAN Aggregation: Another good best practice to follow if SPAN port usage is required, Aggregator TAPs allow
you to take those SPAN and consolidate them into just one or two links (2:1, 8:1, 22:1) . This optimizes and reduces
network complexity.
SPAN Data Diodes: Garland’s Data Diode TAPs are designed to secure SPAN links, ensuring no bidirectional traffic
is sent to monitoring tools.
Diagram 4: SPAN regeneration mode for network TAPs Diagram 5: SPAN aggregation with network TAPs
In a test conducted by Packet Pioneer, Chris Greer set out to see the difference between a data stream captured
on a network TAP versus a SPAN port.
The test connected two PCs to a basic Cisco Catalyst Switch at 100Mbps. A throughput test using iPerf was
configured and run between the two machines. On one of the PCs, he placed a 100Mbps TAP, and a hardware
analyzer to capture. Lastly, he configured a SPAN on the switch to forward all traffic to and from this port to
another hardware analyzer.
The throughput test finished with a result of 93.1Mbps sustained for 10 seconds between the two PCs.
The SPAN data capture showed almost 8,000 packets missing from the trace. This represents almost 8% of the
total packets that were captured by the analyzer from the network TAP. We should also point out that this was on
a 100Mbps interface, not a Gigabit interface, and there were no errored frames. The switch bus was not in a near
overloaded state.
Also, the difference in the timing between the TCP SYN and SYN ACK in the two traces shows us that the switch
is not treating both the SPAN and Destination ports the same. In fact, it was forwarding traffic to the SPAN port
faster than the true destination. While the difference is only 21 uSec, it shows that the switch is affected when
SPAN is enabled. It is not as seamless as it would appear, and this delay was under no load test. With the switch
loaded with traffic, the losses and timing will show greater differential and also dropped packets.
Considering the results of their test, Chris Greer, a network analyst at Packet Pioneer, said, “I am now a full believer
in using a real [network] TAP whenever possible, especially when timing and total view of the data is important!”
TAPs VS SPAN
• Provides 100% full duplex copies of network traffic, • Provides access to packets for monitoring
ensuring no dropped packets for monitoring • SPAN traffic is the lowest priority on the switch
• TAPs are passive or failsafe, ensuring no single point • Switch will drop SPAN packets if heavily utilized
of failure (SPOF) or oversubscribed
• TAPs do not alter the time relationships of frames, • Corrupt packets and low layer errors can be
spacing and response times, which is especially dropped out by a switch/SPAN
important with VoIP and Triple Play analysis • SPAN can duplicate packets if multiple VLANs
including FDX analysis are used
• TAPs do not introduce any additional jitter or • Using SPAN/Mirror ports can change the timing
distortion which is important in VoIP / Video analysis of the frame interactions, altering response times
• VLAN tags are not passed through the SPAN port so • SPAN is not legally compliant for lawful
this can lead to false issues detected and difficulty intercept cases
in finding VLAN issues • SPAN ports can easily be incorrectly configured
• TAPs pass all traffic: IPv4 or IPv6, error packets, impacting network performance, and even
short or large frames, bad CRC frames, interframe cause outages
gap is not dropped nor altered, packets are not • Bidirectional traffic opens back flow of traffic
dropped regardless of the bandwidth into the network, making switch susceptible
• TAPs are fault tolerant to hacking
• TAPs are secure, do not have an IP address or MAC • SPAN retention. SPAN ports are limited in
address, and cannot be hacked number compared to how many are needed for
• CALEA (Commission on Accreditation for Law monitoring and can be costly as available ports
Enforcement Agencies) approved for lawful come at a premium
intercept, providing forensically sound data, • Admin/programming costs for SPAN can get
ensuring 100% accurate data captured with progressively more time intensive and costly
time reference
• Simple, plug and play. TAPs typically have little
setup or command line issues, data is assured and
saves users setup time
• TAPs are timeless - They never need to download or
be upgraded, they do not have access to anything
except the LAN they are monitoring
• Scaleable for traffic optimization, can regenerate
one link to multiple or aggregate multiple links
down to one
afhi95mcqoa01gcejeafeknach/sc_may2016_dg_html/index.html#page/NPM%2520Deployment%2520Guide%2Fpacket_collection.06.3.html%23
2 - Cisco | https://www.cisco.com/c/en/us/support/docs/switches/catalyst-6500-series-switches/10570-41.html