Explicit Flow Measurements Techniques

3.2.1. Generation Phase

Only client is actively involved in the generation phase.¶

When connection starts and spin bit is set to 0, the client initializes the delay bit of the first packet to 1, so it becomes the delay sample for that marking period. Only this packet is marked with the delay bit set to 1 for this round trip period; the other ones will carry the spin bit, while the delay bit will be set to 0.¶

The server initializes the delay bit to 0 at the beginning of the connection, and its only task during the connection is described in Section 3.2.2.¶

In absence of network impairments, the delay sample should bounce between client and server continuously, for the entire duration of the connection. That is highly unlikely for two reasons:¶

1. the packet carrying the delay bit might be lost;¶
2. an endpoint could stop or delay sending packets because the application is limiting the amount of traffic transmitted;¶

To deal with these problems, the algorithm provides a procedure named "recovery process" to regenerate the delay sample and to inform a possible observer of the problem so the measurement can be restarted.¶

3.2.2. Reflection Phase

Reflection is the process that enables the bouncing of the delay sample between a client and a server. The behavior of the two endpoints is slightly different.¶

• Server side reflection: when a delay sample arrives, the server marks the first packet in the opposite direction as the delay sample, if the outgoing packet has the same spin bit value as the delay sample. Otherwise, the delay sample is ignored.¶
• Client side reflection: when a delay sample arrives, the client marks the first packet in the opposite direction as the delay sample, if the outgoing packet has the opposite spin bit value then the delay sample. Otherwise, the delay sample is ignored.¶

In both cases, if the outgoing delay sample is being transmitted with a delay greater than a predetermined threshold after the reception of the incoming delay sample (1ms by default), the delay sample is not reflected, and the outgoing delay bit is kept at 0.¶

Note that reflection takes place for the delay sample regardless of its position within the spin period, as long as it stays within its original spin period.¶

A time threshold for the retransmission of the delay sample is used to eliminate measurements that would overestimate the delay due to lack of traffic on the endpoints. Hence, the maximum estimation error would amount to twice the threshold (e.g. 2ms) per measurement.¶

3.2.3. Two Bits Delay Measurement: Spin Bit + Delay Bit

When the Delay Bit is used, a passive observer can use delay samples directly and avoid inherent ambiguities in the calculation of the RTT in spin bit analysis, such as heuristic validation of the goodness of an edge signal.¶

           =======================|======================>
           = **********     -----Obs---->     ********** =
           = * Client *                       * Server * =
           = **********     <------------     ********** =
           <==============================================

                     (a) client-server RTT

           ==============================================>
           = **********     ------------>     ********** =
           = * Client *                       * Server * =
           = **********     <----Obs-----     ********** =
           <======================|=======================

                     (b) server-client RTT

           =======================>
           = **********     ------|----->     **********
           = * Client *          Obs          * Server *
           = **********     <-----|------     **********
           <=======================

                  (a) client-observer half-RTT

                                  =======================>
             **********     ------|----->     ********** =
             * Client *          Obs          * Server * =
             **********     <-----|------     ********** =
                                  <=======================

                  (b) observer-server half-RTT

        =========================================>
        = =====================>
        = = **********      ---|-->           ---|-->      **********
        = = * Client *         Obs               Obs       * Server *
        = = **********      <--|---           <--|---      **********
        = <=====================
        <=========================================

                 (a) client-observer RTT components (half-RTTs)

                               ==================>
            **********      ---|-->           ---|-->      **********
            * Client *         Obs               Obs       * Server *
            **********      <--|---           <--|---      **********
                               <==================

                 (b) the intra-domain RTT resulting from the
                 subtraction of the above RTT components

3.2.4. Observer's Algorithm and Edge Rejection Interval

To provide a formal description of the observer behavior, we define a "matching delay sample" to be a delay sample with the spin bit value that matched the spin bit value of then-current spin period.¶

Upon detecting a matching delay sample, if a matching delay sample was also detected in the previous period, then the two delay samples can be used to calculate RTT measurement.¶

If the observer can observe both forward and return traffic flows, and it is able to determine which direction contains the client and the server (e.g. by observing the spin bit or connection handshake):¶

• If matching delay samples have been detected in both directions in the current spin period, they can be used to measure the observer-server half-RTT.¶
• If a matching delay sample has been detected in client-to-observer direction, AND a matching delay sample had been detected in observer-to-client direction in the previous spin period, they can be used to measure the observer-client half-RTT.¶

The described observer behavior depends on the ability to accurately identify current spin periods and to reject spurious spin edges, caused by packet reordering. Failure to do so will lead to many missed measurement opportunities and will decrease the amount of usable delay samples available to the observer.¶

To implement spurious edge rejection, every time a spin bit edge is detected, the observer starts a new spin period and begins a time interval during which it rejects spin edges (e.g. 5ms). This guarantees protection against spurious edges due to packets that have been reordered by less than the time interval. The mechanism only works for intervals smaller than the RTT of the observed connection; if RTT is smaller than the edge rejection interval, the observer cannot measure the RTT.¶

4.1.1. Round Trip Packet Loss Measurement

Since the measurements are performed on a portion of the traffic exchanged between the client and the server, the observer calculates the end-to-end Round Trip Packet Loss (RTPL) that, statistically, will correspond to the loss rate experienced by the connection along the entire network path.¶

           =======================|======================>
           = **********     -----Obs---->     ********** =
           = * Client *                       * Server * =
           = **********     <------------     ********** =
           <==============================================

                     (a) client-server RTPL

           ==============================================>
           = **********     ------------>     ********** =
           = * Client *                       * Server * =
           = **********     <----Obs-----     ********** =
           <======================|=======================

                     (b) server-client RTPL

This methodology also allows the Half-RTPL measurement and the Intra-domain RTPL measurement in a way similar to RTT measurement.¶

           =======================>
           = **********     ------|----->     **********
           = * Client *          Obs          * Server *
           = **********     <-----|------     **********
           <=======================

                  (a) client-observer half-RTPL

                                  =======================>
             **********     ------|----->     ********** =
             * Client *          Obs          * Server * =
             **********     <-----|------     ********** =
                                  <=======================

                  (b) observer-server half-RTPL

                           =========================================>
                                             =====================> =
        **********      ---|-->           ---|-->      ********** = =
        * Client *         Obs               Obs       * Server * = =
        **********      <--|---           <--|---      ********** = =
                                             <===================== =
                           <=========================================

             (a) observer-server RTPL components (half-RTPLs)

                           ==================>
        **********      ---|-->           ---|-->      **********
        * Client *         Obs               Obs       * Server *
        **********      <--|---           <--|---      **********
                           <==================

             (b) the intra-domain RTPL resulting from the
             subtraction of the above RTPL components

4.1.2. Setting the Round Trip Loss Bit on Outgoing Packets

The round Trip loss signal requires a working Spin-bit signal to separate trains of marked packets (packets with T bit set to 1). A "pause" of at least one empty spin-bit period between each phase of the algorithm serves as such separator for the on-path observer.¶

The client is in charge of launching trains of marked packets and does so according to the algorithm:¶

1. Generation Phase. The client starts generating marked packets for two consecutive spin-bit periods; it maintains a "generation token" count that is reset to zero at the beginning of the algorithm phase and is incremented every time a packet arrives. When the client transmits a packet and a "generation token" is available, the client marks the packet and retires a "generation token". If no token is available, the outgoing packet is transmitted unmarked. At the end of the first spin-bit period spent in generation, the reflection counter is unlocked to start counting incoming marked packets that will be reflected later;¶
2. Pause Phase. When the generation is completed, the client pauses till it has observed one entire spin bit period with no marked packets. That spin bit period is used by the observer as a separator between generated and reflected packets. During this marking pause, all the outgoing packets are transmitted with T bit set to 0. The reflection counter is still incremented every time a marked packet arrives;¶
3. Reflection Phase. The client starts transmitting marked packets, decrementing the reflection counter for each transmitted marked packet until the reflection counter reached zero. The "generation token" method from the generation phase is used during this phase as well. At the end of the first spin-period spent in reflection, the reflection counter is locked to avoid incoming reflected packets incrementing it;¶
4. Pause Phase 2. The pause phase is repeated after the reflection phase and serves as a separator between the reflected packet train and a new packet train.¶

The generation token counter should be capped to limit the effects of a subsequent sudden reduction in the other endpoint's packet rate that could prevent that endpoint from reflecting collected packets. The most conservative cap value is 1.¶

A server maintains a "marking counter" that starts at zero and is incremented every time a marked packet arrives. When the server transmits a packet and the "marking counter" is positive, the server marks the packet and decrements the "marking counter". If the "marking counter" is zero, the outgoing packet is transmitted unmarked.¶

4.1.3. Observer's Logic for Round Trip Loss Signal

The on-path observer counts marked packets and separates different trains by detecting spin-bit periods (at least one) with no marked packets. The Round Trip Packet Loss (RTPL) is the difference between the size of the Generation train and the Reflection train.¶

In the following example, packets are represented by two bits (first one is the spin bit, second one is the loss bit):¶

        Generation          Pause           Reflection       Pause
   ____________________ ______________ ____________________ ________
  |                    |              |                    |        |
   01 01 00 01 11 10 11 00 00 10 10 10 01 00 01 01 10 11 10 00 00 10

Note that 5 marked packets have been generated of which 4 have been reflected.¶

4.1.4. Loss Coverage and Signal Timing

A cycle of the round Trip loss signaling algorithm contains 2 RTTs of Generation phase, 2 RTTs of Reflection phase, and two Pause phases at least 1 RTT in duration each. Hence, the loss signal is delayed by about 6 RTTs since the loss events.¶

The observer can only detect loss of marked packets that occurs after its initial observation of the Generation phase and before its subsequent observation of the Reflection phase. Hence, if the loss occurs on the path that sends packets at a lower rate (typically ACKs in such asymmetric scenarios), 2/6 (1/3) of the packets will be sampled for loss detection.¶

If the loss occurs on the path that sends packets at a higher rate, lowPacketRate/(3*highPacketRate) of the packets will be sampled for loss detection. For protocols that use ACKs, the portion of packets sampled for loss in the higher rate direction during unidirectional data transfer is 1/(3*packetsPerAck), where the value of packetsPerAck can vary by protocol, by implementation, and by network conditions.¶

4.2.1. Q Block Length Selection

The length of the block must be known to the on-path network probes. There are two alternatives to selecting the Q Block length. The first one requires that the length is known a priori and therefore set within the protocol specifications that implements the marking mechanism. The second requires the sender to select it.¶

In this latter scenario, the sender is expected to choose N (Q Block length) based on the expected amount of loss and reordering on the path. The choice of N strikes a compromise - the observation could become too unreliable in case of packet reordering and/or severe loss if N is too small, while short flows may not yield a useful upstream loss measurement if N is too large (see Section 4.2.2).¶

The value of N should be at least 64 and be a power of 2. This requirement allows an Observer to infer the Q Block length by observing one period of the square signal. It also allows the Observer to identify flows that set the loss bits to arbitrary values (see Section 7).¶

If the sender does not have sufficient information to make an informed decision about Q Block length, the sender should use N=64, since this value has been extensively tried in large-scale field tests and yielded good results. Alternatively, the sender may also choose a random power-of-2 N for each flow, increasing the chances of using a Q Block length that gives the best signal for some flows.¶

The sender must keep the value of N constant for a given flow.¶

4.2.2. Upstream Loss

Blocks of N (Q Block length) consecutive packets are sent with the same value of the Q bit, followed by another block of N packets with an inverted value of the Q bit. Hence, knowing the value of N, an on-path observer can estimate the amount of upstream loss after observing at least N packets. The upstream loss rate (uloss) is one minus the average number of packets in a block of packets with the same Q value (p) divided by N (uloss=1-avg(p)/N).¶

The observer needs to be able to tolerate packet reordering that can blur the edges of the square signal, as explained in Section 4.2.3.¶

          =====================>
          **********     -----Obs---->     **********
          * Client *                       * Server *
          **********     <------------     **********

            (a) in client-server channel (uloss_up)

          **********     ------------>     **********
          * Client *                       * Server *
          **********     <----Obs-----     **********
                               <=====================

            (b) in server-client channel (uloss_down)

4.2.3. Identifying Q Block Boundaries

Packet reordering can produce spurious edges in the square signal. To address this, the observer should look for packets with the current Q bit value up to X packets past the first packet with a reverse Q bit value. The value of X, a "Marking Block Threshold", should be less than N/2.¶

The choice of X represents a trade-off between resiliency to reordering and resiliency to loss. A very large Marking Block Threshold will be able to reconstruct Q Blocks despite a significant amount of reordring, but it may erroneously coalesce packets from multiple Q Blocks into fewer Q Blocks, if loss exceeds 50% for some Q Blocks.¶

4.3.1. End-To-End Loss

The Loss Event bit allows an observer to estimate the end-to-end loss rate by counting packets with L bit value of 0 and 1 for a given flow. The end-to-end loss rate is the fraction of packets with L=1.¶

The assumption here is that upstream loss affects packets with L=0 and L=1 equally. If some loss is caused by tail-drop in a network device, this may be a simplification. If the sender's congestion controller reduces the packet send rate after loss, there may be a sufficient delay before sending packets with L=1 that they have a greater chance of arriving at the observer.¶

4.3.2. Loss Profile Characterization

In addition to measuring the end-to-end loss rate, the Loss Event bit allows an observer to characterize loss profile, since the distribution of observed packets with L bit set to 1 roughly corresponds to the distribution of packets lost between 1 RTT and 1 RTO before (see Section 4.4.1). Hence, observing random single instances of L bit set to 1 indicates random single packet loss, while observing blocks of packets with L bit set to 1 indicates loss affecting entire blocks of packets.¶

4.4.1. Correlating End-to-End and Upstream Loss

Upstream loss is calculated by observing packets that did not suffer the upstream loss (Section 4.2.2). End-to-end loss, however, is calculated by observing subsequent packets after the sender's protocol detected the loss. Hence, end-to-end loss is generally observed with a delay of between 1 RTT (loss declared due to multiple duplicate acknowledgments) and 1 RTO (loss declared due to a timeout) relative to the upstream loss.¶

The flow RTT can sometimes be estimated by timing protocol handshake messages. This RTT estimate can be greatly improved by observing a dedicated protocol mechanism for conveying RTT information, such as the Spin bit (see Section 3.1) or Delay bit (see Section 3.2).¶

Whenever the observer needs to perform a computation that uses both upstream and end-to-end loss rate measurements, it should use upstream loss rate leading the end-to-end loss rate by approximately 1 RTT. If the observer is unable to estimate RTT of the flow, it should accumulate loss measurements over time periods of at least 4 times the typical RTT for the observed flows.¶

If the calculated upstream loss rate exceeds the end-to-end loss rate calculated in Section 4.3.1, then either the Q Period is too short for the amount of packet reordering or there is observer loss, described in Section 4.4.1.2. If this happens, the observer should adjust the calculated upstream loss rate to match end-to-end loss rate, unless the following applies.¶

In case of a protocol like TCP and SCTP that does not track losses of pure ACK packets, observing a direction of traffic dominated by pure ACK packets could result in measured upstream loss that is higher than measured end-to-end loss, if said pure ACK packets are lost upstream. Hence, if the measurement is applied to such protocols, and the observer can confirm that pure ACK packets dominate the observed traffic direction, the observer should adjust the calculated end-to-end loss rate to match upstream loss rate.¶

4.5.1. R+Q Bits - Using R and Q Bits for Passive Loss Measurement

Since both sQuare and Reflection square bits are toggled at most every N packets (except for the first transition of the R-bit as explained before), an on-path observer can count the number of packets of each marking block and, knowing the value of N, can estimate the amount of loss experienced by the connection. An observer can calculate different measurements depending on whether it is able to observe a single direction of the traffic or both directions.¶

Single directional observer:¶

• upstream loss in the observed direction: the loss between the sender and the observation point (see Section 4.2.2)¶
• "three-quarters" connection loss: the loss between the receiver and the sender in the unobserved direction plus the loss between the sender and the observation point in the observed direction¶
• end-to-end loss in the unobserved direction: the loss between the receiver and the sender in the opposite direction¶

Two directions observer (same metrics seen previously applied to both direction, plus):¶

• client-observer half round-trip loss: the loss between the client and the observation point in both directions¶
• observer-server half round-trip loss: the loss between the observation point and the server in both directions¶
• downstream loss: the loss between the observation point and the receiver (applicable to both directions)¶

        =======================>
        = **********     -----Obs---->     **********
        = * Client *                       * Server *
        = **********     <------------     **********
        <============================================

            (a) in client-server channel (tqloss_up)

          ============================================>
          **********     ------------>     ********** =
          * Client *                       * Server * =
          **********     <----Obs-----     ********** =
                               <=======================

            (b) in server-client channel (tqloss_down)

          **********     -----Obs---->     **********
          * Client *                       * Server *
          **********     <------------     **********
          <==========================================

            (a) in client-server channel (eloss_down)

          ==========================================>
          **********     ------------>     **********
          * Client *                       * Server *
          **********     <----Obs-----     **********

            (b) in server-client channel (eloss_up)

        =======================>
        = **********     ------|----->     **********
        = * Client *          Obs          * Server *
        = **********     <-----|------     **********
        <=======================

      (a) client-observer half round-trip loss (hrtloss_co)

                               =======================>
          **********     ------|----->     ********** =
          * Client *          Obs          * Server * =
          **********     <-----|------     ********** =
                               <=======================

      (b) observer-server half round-trip loss (hrtloss_os)

                               =====================>
          **********     ------|----->     **********
          * Client *          Obs          * Server *
          **********     <-----|------     **********

             (a) in client-server channel (dloss_up)

          **********     ------|----->     **********
          * Client *          Obs          * Server *
          **********     <-----|------     **********
          <=====================

             (b) in server-client channel (dloss_down)

4.5.2. Enhancement of R Block Length Computation

The use of the rounding function used in the M computation introduces errors that can be minimized by storing the rounding applied each time M is computed, and using it during the computation of the M value in the following R Block.¶

This can be achieved introducing the new r_avg parameter in the computation of M. The new formula is Mr=avg(p)+r_avg; M=round(Mr); r_avg=Mr-M where the initial value of r_avg is equal to 0.¶

4.5.3. Improved Resilience to Packet Reordering

When a protocol implementing the marking mechanism is able to detect when packets are received out of order, it can improve resilience to packet reordering beyond what is possible using methods described in Section 4.2.3.¶

This can be achieved by updating the size of the current R Block while this is being transmitted. The reflection block size is then updated every time an incoming reordered packet of the previous Q Block is detected. This can be done if and only if the transmission of the current reflection block is in progress and no packets of the following Q Block have been received.¶