Ygdal Naouri et al.

Patent Trials and Appeals Board

Dec 27, 2019

14285028 - (D) (P.T.A.B. Dec. 27, 2019)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
14/285,028 05/22/2014 Ygdal Naouri P62688 2072
96162 7590 12/27/2019
Law Office of R. Alan Burnett, PS
c/o CPA Global
900 Second Avenue South, Suite 600
Minneapolis, MN 55402
EXAMINER
MIAN, OMER S
ART UNIT PAPER NUMBER
2461
NOTIFICATION DATE DELIVERY MODE
12/27/2019 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
alan@patentlylegal.com
docketing@cpaglobal.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte YGDAL NAOURI, ROBERT O. SHARP, KENNETH G. KEELS,
and ERIC W. MULTANEN
Appeal 2018-007825
Application 14/285,028
Technology Center 2400
Before ROBERT E. NAPPI, JOHNNY A. KUMAR, and
LINZY T. McCARTNEY, Administrative Patent Judges.
McCARTNEY, Administrative Patent Judge.
DECISION ON APPEAL
Appellant1 seeks review under 35 U.S.C. § 134(a) of the Examiner’s
final rejection of claims 1–19 and 21–26. We have jurisdiction under
35 U.S.C. § 6(b).
We affirm in part.

1 Appellant identifies the real party in interest as Intel Corporation. Appeal
Brief 3, filed April 13, 2018 (“Appeal Br.”).
Appeal 2018-007825
Application 14/285,028

2
BACKGROUND
This patent application concerns data center congestion management
for non-Transmission Control Protocol traffic. See, e.g., Specification ¶ 17,
filed May 22, 2014 (“Spec.”). Claims 1, 11, and 18 are independent. Claim 1
illustrates the claimed subject matter:
1. A method for implementing congestion management of
non-TCP (Transmission Control Protocol) traffic in a network,
comprising:
determining a non-congested transit latency for a path
between a source end-node and a destination end-node, wherein
the path traverses at least one switch and the non-congested
transit latency is an overall transit latency for the path;
measuring transit latencies for transfer of non-TCP
packets or Ethernet frames encapsulating non-TCP packets along
the path, each transit latency comprising an overall transit latency
for the path;
determining, for transfer of each of at least a portion of the
non-TCP packets or Ethernet frames, whether to mark the path
as congested or not congested based on a difference between the
transit latency measured for the non-TCP packet or Ethernet
frame and the non-congested transit latency determined for the
path; and
managing a rate at which the non-TCP packets are
transmitted from the source end-node to be forwarded via the
path to the destination end-node based as a function of a rate at
which the path is marked as congested.
Appeal Br. 44.

Appeal 2018-007825
Application 14/285,028

3
REJECTIONS2
Claims 35 U.S.C. § References
1, 3–9, 11, 18, 21,
26 103 Cafiero,
3 Dahlin4
2, 12, 19 103 Cafiero, Dahlin, Attar5
10, 13–17, 22–25 103 Cafiero, Dahlin, Corlett6
DISCUSSION
We have reviewed the Examiner’s rejections and Appellant’s
arguments, and with the exception discussed below, we disagree with
Appellant that the Examiner erred. As consistent with the discussion below,
we adopt the Examiner’s reasoning, findings, and conclusions on pages 3–38
of the Final Office Action, pages 4–15 of the Advisory Action mailed
February 5, 2018 (“Advisory Act.”), and pages 2–34 of the Examiner’s
Answer. We address Appellant’s arguments in turn.
Claim 1
“Determining a Non-Congested Transit Latency”
Claim 1 recites “determining a non-congested transit latency for a
path between a source end-node and a destination end-node, wherein the
path traverses at least one switch and the non-congested transit latency is an
overall transit latency for the path.” Appeal Br. 44. Appellant contends that

2 In the Final Office Action, the Examiner rejected the pending claims under
35 U.S.C. § 112(b). Final Office Action 2, mailed November 16, 2017
(“Final Act.”). The Examiner withdrew this rejection in the Examiner’s
Answer. Examiner’s Answer 2, mailed May 29, 2018 (“Ans.”).
3 Cafiero et al. (US 2006/0098681 A1; May 11, 2006).
4 Dahlin et al. (US 2004/0064577 A1; April 1, 2004).
5 Attar et al. (US 2011/0211449 A1; September 1, 2011).
6 Corlett (US 2009/0161569 A1; June 25, 2009).
Appeal 2018-007825
Application 14/285,028

4
one of ordinary skill in the art would “not need to combine Cafiero with
Dahlin to obtain [the recited] non-congested overall latency” apparently
because Dahlin “on its own, can determine the non-congested transit latency
for a round-trip by determining the minRTT” or minimum observed round
trip time. Reply Brief 8, filed July 29, 2018 (“Reply Brief”) (emphases
omitted).
We find this argument unpersuasive. Although one of ordinary skill in
the art may not have needed to modify Dahlin to include the recited non-
congested transit latency, the Examiner found that Cafiero does not teach
aspects of this latency. See, e.g., Final Act 5, 29. The Examiner found that
Cafiero discloses determining an age limit for latency control (referred to as
TL) that corresponds to the recited “determining a non-congested transit
latency.” See, e.g., Final Act. 4. But the Examiner found that Cafiero does
not teach that this age limit “is an overall transit latency for the path” as
required by claim 1. See Final Act. 5, 29. To remedy this deficiency, the
Examiner turned to Dahlin. See Final Act. 5–6, 29–30. The Examiner found
that Dahlin discloses a minimum round trip time (minRTT) that teaches a
non-congested transit latency that is an overall transmit latency for a path.
See Final Act. 5, 29–30; see also Dahlin ¶ 15 (“An estimate of uncongested
round trip time may be based on a minimum round trip time for a data
packet that has been detected ( e.g., within a specific time period).”).
Appellant acknowledges that Dahlin’s minimum round trip time is a non-
congested transit latency. See Appeal Br. 30 (conceding that Dahlin’s
“minRTT could be used as a non-congested transit latency”); Reply Br. 8
(admitting that Dahlin “can determine the non-congested transit latency for a
round-trip by determining the minRTT”).
Appeal 2018-007825
Application 14/285,028

5
The Examiner concluded that it would have been obvious to replace
Cafiero’s age limit for latency control with a minimum roundtrip time to
arrive at the claimed invention because doing so would provide more
accurate latency measurements, among other advantages. See Ans. 25
(explaining that using a roundtrip time “gives a more accurate and not just
an inferred measurement by giving a more direct determination of latency in
terms of time in both direction”). We thus find unpersuasive Appellant’s
contention that one of skill in the art would not have needed to combine
Cafiero with Dahlin.
“Measuring Transit Latencies”
Claim 1 also recites “measuring transit latencies for transfer of non-
TCP packets or Ethernet frames encapsulating non-TCP packets along the
path, each transit latency comprising an overall transit latency for the path.”
Appeal Br. 44. Appellant argues that Dahlin cannot teach measuring transit
latencies for transfer of non-TCP packets or Ethernet frames encapsulating
non-TCP packets as recited in claim 1 because Dahlin uses TCP to measure
roundtrip times. See Appeal Br. 27, 30–31. For this reason, Appellant also
contends that “it would not be possible” to combine Dahlin’s and Cafiero’s
inventions to “measur[e] transit latencies for transfer of non-TCP packets or
Ethernet frames encapsulating non-TCP packets along the path, each transit
latency comprising an overall transit latency for the path.” Appeal Br. 32;
see also Appeal Br. 34.
We find these arguments unpersuasive. The Examiner found Cafiero
discloses packet ages that correspond to the recited “measuring transit
latencies for transfer of non-TCP packets or Ethernet frames encapsulating
non-TCP packets along the path” but does not disclose the each transit
Appeal 2018-007825
Application 14/285,028

6
latency “compris[es] an overall transit latency for the path” as required by
claim 1. See, e.g., Final Act. 4, 28–29. But the Examiner found that Dahlin
discloses a roundtrip time (RTT) that teaches transit latencies that comprise
an overall transit latency for the path and concluded that it would have been
obvious to replace Cafiero’s packet ages with a round trip time to arrive at
the claimed invention. See, e.g., Final Act. 5–6; Ans. 9–10.
Contrary to Appellant’s arguments, the Examiner did not conclude
that it would have been obvious to incorporate Dahlin’s use of TCP into
Cafiero. As discussed above, the Examiner relied on Dahlin only for the
concept of roundtrip times. See Final Act. 32 (explaining that the Examiner
used Dahlin “only . . . to teach . . . latency measurement[s] based on round
trip times” and did not modify Cafiero by “the entire invention of” Dahlin);
Ans. 28 (explaining that Dahlin “is only relied upon to teach round trip
times” (emphasis modified)). In any event, “[t]he test for obviousness is not
whether the features of a secondary reference may be bodily incorporated
into the structure of the primary reference . . . . Rather, the test is what the
combined teachings of the references would have suggested to those of
ordinary skill in the art.” In re Keller, 642 F.2d 413, 425 (CCPA 1981).
Thus, Appellant’s arguments that one of ordinary skill in the art could not
have combined Cafiero’s and Dahlin’s inventions have not shown that the
Examiner erred. Cf. In re Etter, 756 F.2d 852, 859 (Fed.Cir.1985) (en banc)
(“Etter’s assertions that Azure cannot be incorporated in Ambrosio are
basically irrelevant, the criterion being not whether the references could be
physically combined but whether the claimed inventions are rendered
obvious by the teachings of the prior art as a whole.”).
Appeal 2018-007825
Application 14/285,028

7
The Examiner also found that the roundtrip times disclosed by Dahlin
“are generic to all packet based communication protocols” and
“implementable in any packet[-]base[d] protocol.” Ans. 28; see also Final
Act. 32 (finding that “[d]etermining round-trip times of packets traversing a
path in a communication networks is a well-known technique in the art”).
Appellant contends that the Examiner “cites no factual basis or evidence for”
these findings and has not shown how one of ordinary skill in the art could
modify Cafiero with Dahlin’s teachings “to determine a round trip time
without using TCP.” Reply Br. 10; see also Reply Br. 20.
We find these arguments unpersuasive. Dahlin discloses that “[a]
round trip time may refer to an elapsed time between the time that a data
packet is sent and the time that the acknowledgement of receipt of the data
packet is received.” Dahlin ¶ 14. Thus, to determine these roundtrip times in
Cafiero’s system, the system would have to send and receive packets,
determine when the packets were sent and received, and determine
differences in time. Cafiero teaches sending and receiving packets,
timestamping packets when certain events occur (for example, when the
packet enters a buffer), and determining differences between a timestamp
and the current time. See, e.g., Cafiero ¶¶ 6 (explaining that the words
“packet” and “frame” are used interchangeably in Cafiero), 73–75
(discussing sending packets and receiving frames), 142 (discussing
timestamping packets), 143 (discussing comparing a timestamp with a
current time). Given these disclosures, we see no error in the Examiner’s
finding that those of ordinary skill in the art would have been able to
measure roundtrip times in Cafiero. We also note that another reference,
Appeal 2018-007825
Application 14/285,028

8
Corlett, teaches calculating roundtrip times. See, e.g., Corlett ¶¶ 15, 45, 98–
100, Abstract.
“Determining . . . Whether to Mark the Path”
Claim 1 also recites “determining, for transfer of each of at least a
portion of the non-TCP packets or Ethernet frames, whether to mark the path
as congested or not congested based on a difference between the transit
latency measured for the non-TCP packet or Ethernet frame and the non-
congested transit latency determined for the path.” Appeal Br. 44. Appellant
argues that the Examiner’s combination of Cafiero and Dahlin does not
teach or suggest this limitation because the combination would result in
comparing Cafiero’s age value with Dahlin’s minRTT measurement, which
Appellant contends are “completely different measurements.” Appeal Br. 33.
We disagree. As discussed above, the Examiner concluded that it
would have been obvious to replace Cafiero’s age for a packet with a round
trip time and age limit for latency control with a minimum round trip time.
See, e.g., Final Act. 29–30; Ans. 5–10. Thus, the Examiner’s combination of
Cafiero and Dahlin would result in comparing a round trip time with a
minimum round trip time, not an age value with a minimum round trip time.
We thus find this argument unpersuasive.
Motivation to Combine and Reasonable Expectation of Success
Appellant argues that the “the Examiner’s motivation to combine has
nothing to do with the claimed” invention because the context of the
invention generally involves “use in a data center or the like” and the
Examiner’s motivation concerns “the use of a self-tuning background
replication layer.” Appeal Br. 34, 35. Appellant asserts that this use would
“greatly limit the usefulness of the combination, and it could not be applied
Appeal 2018-007825
Application 14/285,028

9
to managing data center traffic congestion on a general level.” Appeal Br.
35.
We disagree. As acknowledged by Appellant, claim 1 is “not limited
exclusively to . . . use in a data center or the like,” Appeal Br. 34, and
Appellant has provided no persuasive evidence or reasoning to support the
assertion that using a self-tuning background replication layer would
“greatly limit the usefulness of the combination” and “could not be applied
to managing data center traffic congestion on a general level,” Appeal Br.
35. More important, Appellant ignores that the Examiner provided other
reasons to combine Cafiero’s and Dahlin’s teachings in the claimed manner.
The Examiner found that the proposed combination would provide more
accurate latency measurements and better information about the status of
network traffic. See Ans. 25 (explaining that using a round trip time “gives a
more accurate and not just an inferred measurement by giving a more direct
determination of latency in terms of time in both direction[s]”); Final Act. 30
(explaining that using round trip times “provides a well-known technique of
measuring the round-trip latency as opposed to a latency towards the
destination only to give a better overall status of the network traffic in a two-
way communication service”). This reasoning provides an adequate
motivation to combine the teachings of the cited art in the claimed manner.
Appellant also argues there would have been no motivation to
combine the teachings of Cafiero and Dahlin or a reasonable expectation of
success in doing so because the references “involve totally different
approaches to managing network congestion” and “it would be impossible
for the proposed combination to meet the limitations” of claim 1. Appeal Br.
34, 35; see also Appeal Br. 33, 36. According to Appellant, combining these
Appeal 2018-007825
Application 14/285,028

10
different approaches to managing network congestion would make Cafiero
inoperable because Cafiero’s “use of buffer fill levels and ECN would be
replaced.” Appeal Br. 35.
We again disagree. Appellant’s arguments assume that the Examiner’s
combination involves every aspect of Cafiero’s and Dahlin’s inventions. The
Examiner repeatedly made clear that it does not. See Final Act. 32
(explaining that the Examiner used Dahlin “only . . . to teach . . . latency
measurement[s] based on round trip times” and did not modify Cafiero by
“the entire invention of” Dahlin); Ans. 28 (explaining that Dahlin “is only
relied upon to teach round trip times and at least one measurement of round
trip time being ‘non-congested transit latency’” (emphasis modified)). And
as explained above, those of ordinary skill in the art would have been able to
determine roundtrip times in Cafiero’s system and thus would have a
reasonable expectation of success in making the Examiner’s proposed
modifications. As for Cafiero’s buffer fill levels and ECN, the Examiner’s
combination does not involve replacing these elements, and Appellant has
provided no persuasive evidence or reasoning that using a roundtrip time or
a minimum roundtrip time in Cafiero’s invention would require these
modifications.
Finally, Appellant contends that one of ordinary skill in the art would
not have looked to Dahlin because the claimed invention manages
congestion of non-TCP traffic and Dahlin uses TCP. Appeal Br. 35. We
understand Appellant to argue that Dahlin is not analogous to the claimed
invention. For a reference to be analogous to a claimed invention, the
reference must be “from the same field of endeavor, regardless of the
problem addressed” or “reasonably pertinent to the particular problem with
Appeal 2018-007825
Application 14/285,028

11
which the inventor is involved.” In re Bigio, 381 F.3d 1320, 1325 (Fed. Cir.
2004). Here, Dahlin is at least reasonably pertinent to the particular problem
with which the inventor is involved. The particular problem faced by the
inventor included developing improved data center congestion management
for non-TCP traffic. See Spec. ¶ 17. Although Dahlin addresses improving
TCP congestion management, see, e.g., Dahlin ¶ 9, one of ordinary skill in
the art looking to improve the congestion control of non-TCP traffic would
have considered how other transmission protocols addressed congestion
control.
In any event, the problem faced by the inventor also included
providing the benefits of TCP congestion control to non-TCP traffic. See,
e.g., Spec. ¶ 3 (“TCP traffic [benefits] from a built-in end-to-end congestion
management method that has been enhanced by several techniques. Data
Center Transport Control Protocol (DCTCP) is the most recent and the most
efficient congestion avoidance variant . . . . Unfortunately, DCTCP is not
relevant for non-TCP traffic . . . .”), 19 (“In accordance with one aspect, the
techniques are implemented in a manner that mimics the DCTCP algorithm
for non-TCP protocols, which has been proven to provide an efficient end-
to-end congestion management method.”). Dahlin addresses modifying a
TCP congestion control protocol to, among other things, “be more sensitive
to congestion than traditional protocols.” Dahlin ¶ 9. Because Dahlin
provides an improved TCP congestion control protocol, Dahlin would have
been reasonably pertinent to the problem of providing the benefits of TCP
congestion control to non-TCP traffic.
Appeal 2018-007825
Application 14/285,028

12
We have considered Appellant’s remaining arguments and find them
unpersuasive. For at least the above reasons, we find Appellant’s arguments
unpersuasive and therefore sustain the Examiner’s rejection of claim 1.
Claim 2
Claim 2 recites “inputting into a DCTCP algorithm a congestion
marking status of the path in place of using a congestion marking status
conveyed via an ECN-Echo flag as an input to the DCTCP algorithm.”
Appeal Br. 44. Appellant argues that the Examiner’s combination of Cafiero
and Dahlin does not teach or suggest this limitation because Cafiero employs
ECN for virtual lanes, not paths. See Appeal Br. 41. Appellant also argues
that the Examiner did not address “replacing the ECN-Echo flag with
anything, much less the congestion marking status of the path” as required
by claim 2. Appeal Br. 42.
Appellant has not persuaded us that the Examiner erred. The
Examiner found Cafiero uses ECN for paths because Cafiero discloses that a
packet age can “indicate[] that the path towards the destination of the packet
is subject to congestion” and that ECN is “considered a core-to-end
notification, because it is sent by core device 1420 and received by NIC card
1465 of end-node 1440.” Ans. 30 (quoting Cafiero ¶¶ 143, 137 (emphases
modified)), 31 (showing an annotated version of Figure 14 of Cafiero). As
for replacing the ECN-Echo flag, the Examiner found Cafiero teaches this
aspect of claim 2 because Cafiero discloses setting the backward ECN
portion (BECN) bit to indicate congestion. See Ans. 32 (quoting Cafiero
¶ 75). Appellant has not explicitly addressed these findings. We therefore
find Appellant’s arguments unpersuasive and thus sustain the Examiner’s
rejection of claim 2.
Appeal 2018-007825
Application 14/285,028

13
Claims 3 and 4
Claim 3 recites “wherein the non-congested transit latency is an
overall transit latency that is measured along a one-way path from the source
end-node to the destination end-node, and each of the transit latencies for the
transfer of the non-TCP packets or the Ethernet frames is an overall transit
latency that is measured along the same one-way path.” Appeal Br. 45.
Appellant argues the parts of Dahlin cited by the Examiner do not teach or
suggest this limitation because they discuss measuring rounding trip times
and make “no reference to measurement of a one-way path.” Appeal Br. 39.
Appellant contends that one of ordinary skill in the art “would not interpret a
measurement of a round trip time that did not separately measure the one
way path from the source node to a destination node to teach this limitation.”
Reply Br. 22; see also Appeal Br. 30.
We agree with Appellant. The Examiner found that Dahlin teaches
this limitation because Dahlin discloses measuring a minimum round trip
time. See Final Act. 9 (citing Dahlin ¶¶ 15, 118); Ans. 27 (citing Dahlin
¶¶ 118–119). The Examiner reasoned that “a roundtrip path includes a one-
way path” thus “any latency value determined for the round trip would be
for a one-way path” and regardless “any latency value measured along the
roundtrip path has to be measured along at least the one-way path.” Ans. 27
(emphases omitted). But as argued by Appellant, one of ordinary skill in the
art would not interpret a measurement of roundtrip time as a latency that is
“measured along a one-way path from the source end-node to the destination
end-node” when the roundtrip time measurement does not include a
separately measured one-way time. See, e.g., Spec. ¶¶ 31–32 (discussing
Appeal 2018-007825
Application 14/285,028

14
one-way and roundtrip time measurements). We therefore do not sustain the
Examiner’s rejection of claim 3 and claim 4, which depends from claim 3.
Claim 11
Claim 11 recites the following:
A server apparatus, configured to be implemented as an
end-node in a data center, the server apparatus having a processor
operatively coupled to memory and operatively coupled to a
network interface configured to support network communication
using an Ethernet protocol, the apparatus further configured,
upon operation, to:
facilitate operation of the server apparatus as a source end-
node;
. . . .
retrieve path congestion marking indicia received from the
destination end-node for transfer of at least a portion of the
Ethernet frames encapsulating non-TCP packets, the path
congestion marking indicia for a given Ethernet frame
identifying whether or not the source destination path was
marked as congested for transfer of the Ethernet frame based on
a difference between a transit latency for the source-destination
path measured for the Ethernet frame by the destination end-
node, and a non-congested transit latency determined for the
source-destination path, each of the transit latency measured for
the Ethernet frame by the destination end-node and the non-
congested transit latency comprising an overall transit latency for
the source-destination path; and
manage a rate at which the non-TCP packets are
transmitted outbound from the transmit port to be forwarded via
the source-destination path to the destination end-node based as
a function of a rate at which the path is marked as congested.
Appeal Br. 47–48.
Appellant argues that the Examiner’s rejection of claim 11 does not
address these limitations. See Appeal Br. 36–37. Appellant also contends
Appeal 2018-007825
Application 14/285,028

15
that Cafiero does not teach or suggest these limitations because Cafiero
performs its operations at “a switch, not a server that is operated as a source
end-node.” Appeal Br. 37. Finally, Appellant asserts that the Examiner erred
in rejecting claim 11 for the same reasons that the Examiner erred in
rejecting claim 1. See Appeal Br. 36.
Appellant has not persuaded us that the Examiner erred. The
Examiner addressed these limitations in an Advisory Action mailed after the
Final Office Action issued. See, e.g., Advisory Action 10–14, mailed
February 5, 2018. Appellant has not explicitly addressed the Examiner’s
findings in the Advisory Action for claim 11 and has therefore not persuaded
us that the Examiner erred. See Manual of Patent Examiner Procedure
§ 1205.02 (“An appellant’s brief must present arguments responsive to every
ground of rejection stated by the examiner in the Office action from which
the appeal has been taken (as modified by any advisory action and/or pre-
appeal brief conference decision).” (emphasis added)). And for the reasons
discussed above, we find Appellant’s arguments about claim 1 (which
Appellant asserts also apply to claim 11) unpersuasive. Accordingly, we
sustain the Examiner’s rejection of claim 11.
Remaining Claims
Appellant does not present separate, persuasive arguments for the
other claims that depend from claim 1 (claims 5–10), the claims that
dependent from independent claim 11 (claims 12–17), or independent claim
18 and its dependent claims (claims 19–26). We therefore sustain the
Examiner’s rejections of these claims.

Appeal 2018-007825
Application 14/285,028

16
CONCLUSION
Claims
Rejected 35 U.S.C. § References Affirmed Reversed
1, 3–9, 11,
18, 21, 26 103 Cafiero, Dahlin
1, 5–9, 11,
18, 21, 26 3, 4
2, 12, 19 103 Cafiero, Dahlin, Attar 2, 12, 19
10, 13–17,
22–25 103 Cafiero, Dahlin, Corlett
10, 13–17,
22–25
Overall
Outcome
1, 2, 5–19,
21–26 3, 4

No period for taking any subsequent action in connection with this
appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(1)(iv).
AFFIRMED IN PART