Subscribe to Test & Measurement World
RSS
Email
Average Rating:
  • (5)
    Rate this:
  • EEEk!: Energy Efficient Ethernet Testing Challenges

    June 3, 2011

    This week, we’ll examine some of the challenges of testing Energy Efficient Ethernet (EEE), a protocol that addresses the growing need for companies and enterprises to be both energy efficient and energy conscious. The EEE protocol, which is defined for most twisted pair PHYs from 10BASE-T through 10GBASE-T, relies on entering an electrically silent low-power idle state when there is no traffic to send, rather than sourcing the IDLE code groups for the given speed. Since the PHYs are entering long periods with no signal, this state adds a great deal of complexity to the monitoring and measurement of their physical signaling characteristics. We’ll look at four of the most common issues of testing EEE devices.

    Establishing Link / Auto-negotiation

    Testing Ethernet PHYs previously depended heavily on test modes, which are special modes of operation where the PHY would continuously transmit particular sequences. However, in addition to these old test modes, many of the most interesting signaling and timers that should be tested during EEE operation can only be observed in a fully established link. Unlike previous electrical testing, which could be performed using only an oscilloscope and the device under test (DUT), this testing requires a link partner (LP) or special test system to establish link while being monitored with another device. Part of that link establishment is performed via Auto-Negotiation which relies on 10BASE-T-like signaling to determine the capabilities of the two devices and resolve to an agreed upon speed. This means that the LP or test system needs to not only be fully capable of EEE signaling, but also of this Auto-Negotiation process.

    Line-tapping / Monitoring

    After link is established, the signaling sent by the DUT still needs to be checked for timing and electrical requirements while switching to and from the low-power idle state. This is done easily enough for 100BASE-TX EEE PHYs by separating the transmit and receive pairs, but for 1000BASE-T and higher speeds, the process is complicated by the fact that all four pairs pass traffic bi-directionally. One solution to this problem is to use a passive bi-directional line tap to non-invasively observe the link on an oscilloscope. In order to use a line-tap in this manner, it needs to be carefully characterized so that its effects can be removed from the measurements during post-processing.

    Decoding 1000BASE-T or Access to Internal Timers

    Another major challenge is that many of the signals that govern the timing when entering and exiting the low power idle state are either internal to the PHY, or encoded into the complex multi-level signaling of the higher speed PHYs. While those truly internal signals can often only be accessed on special development boards, and not on production level equipment, the signals embedded into the signaling are decodable with the right setup. Two methods can be used - a custom designed piece of test equipment that breaks out the necessary signals, or post processing of a capture taken with an oscilloscope. Using the second method, it is possible to fully decode an LPI capture and observe when these signals are sent.  Then, from those points, a test engineer can measure when certain timers are met or when signaling thresholds are crossed.

    Traffic Generation during State of Sleep

    The final challenge we’ll discuss is the synchronization between the generation of traffic and capturing that traffic when using a passive line tap. Since measurements made using a line tap are recorded on an oscilloscope, it is necessary for the link partner or traffic generating device to be capable of sending frames, and it must also be capable of sourcing a trigger signal for the oscilloscope. This can be problematic when trying to use off-the-shelf parts instead of specialized testing equipment, as many devices provide no interface to indicate frame transmission.

    For More Information

    For a more in-depth look at the testing challenges associated with EEE, take a look at the article, “Energy Efficient Ethernet Brings on Testing Challenges,” which we wrote for Test & Measurement World. Here’s a link to a PDF of the article: http://www.tmworld.com/file/25643-Energy_Efficient_Ethernet_UNH_IOL.pdf.

    If you’re interested in testing an EEE-capable device, we will be hosting the Ethernet Alliance’s Energy Efficient Ethernet Interoperability Test Event the second week of August. If you’d like to join us for the test event, visit the event’s website: http://www.ethernetalliance.org/interoperability/ieee_std_8023aztm2010_energy_efficient_ethernet_interoperability_test_event. Or, if you’d like more information, please contact me at jrlapak@iol.unh.edu.

    jeff_lapak_100×100.jpg

    Jeff Lapak - Senior Manager, Ethernet

    Posted by UNH-IOL Staff on June 3, 2011 | Comments (2)
    Industries: Communications Test
    Average Rating:
  • (5)
    Rate this:

  • August 21, 2011
    In response to: EEEk!: Energy Efficient Ethernet Testing Challenges
    Jodi commented:

    Gosh, I wish I would have had that information earleir!


    August 20, 2011
    In response to: EEEk!: Energy Efficient Ethernet Testing Challenges
    Ethanael commented:

    Yeah that's what I'm takling about baby--nice work!

    POST A COMMENT
    Display Name
    captcha

    Before submitting this form, please type the characters displayed above. Note the letters are case sensitive:

    Advertisement
    Advertisement
    Advertisement
    About Us   |   Advertising Info   |   Site Map   |   Contact Us   |   FREE Subscription
    © 2011 UBM Electronics . All rights reserved.
    Use of this Web site is subject to its Terms of Use | Privacy Policy

    Feedback Form
    Feedback Analytics