Ex Parte Zhang et al

Patent Trial and Appeal Board

Jan 24, 2018

14068658 (P.T.A.B. Jan. 24, 2018)

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.
14/068,658 10/31/2013 Ronghua Zhang N056 (NCRA.P0150) 5710
109858 7590
VMware, Inc.
3401 Hill view Avenue
Palo Alto, CA 94304
01/26/2018 EXAMINER
RIVAS, SALVADOR E
ART UNIT PAPER NUMBER
2479
NOTIFICATION DATE DELIVERY MODE
01/26/2018 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):
ipadmin@vmware.com
ipteam @ vmware. com
mail@ adelillp.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte RONGHUA ZHANG and JESSE E. GROSS IV
Appeal 2017-008134
Application 14/068,658
Technology Center 2400
Before DENISE M. POTHIER, SCOTT B. HOWARD, and
STEVEN M. AMUNDSON, Administrative Patent Judges.
POTHIER, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Appellants1 appeal under 35 U.S.C. § 134(a) from the Examiner’s
rejection of claims 1, 3—16, and 19-22. App. Br. I.2 Claim 2 has been
canceled {id. at 28), and claims 17 and 18 are objected to but would be
allowable if rewritten in independent form to include all of the limitations of
1 The real party in interest is listed as Nicira, Inc. App. Br. 2.
2 Throughout this opinion, we refer to (1) the Final Action (Final Act.)
mailed May 5, 2016, (2) the Appeal Brief (App. Br.) filed November 22,
2016, (3) the Examiner’s Answer (Ans.) mailed March 10, 2017, and (4) the
Reply Brief (Reply Br.) filed May 8, 2017.
Appeal 2017-008134
Application 14/068,658
their base claim and any intervening claims (Final Act. 26). We have
jurisdiction under 35 U.S.C. § 6(b). We reverse.
Invention
Appellants’ invention relates to optimizing managing tenant networks
that transmit substantially more data outside the managed network than these
networks receive. See Spec. 12. For example, a managed network manages
forwarding elements operating on host machines to “send packets out onto
the external network through a direct connection that bypasses the
gateways,” which in some embodiments involves “a specific logical port
(called a direct host return (‘DHR’) port) of a logical forwarding element
implemented by the managed forwarding elements.” Id. 13; see also id.
11 19-20, 26,30, Fig. 1.
Claim 1 is reproduced below with emphasis:
1. A method for managing a network for a network controller,
the method comprising:
configuring a first managed forwarding element in the
network, operating in a host machine that hosts a virtual
machine belonging to a particular logical network and
connected to the managed network through the first managed
forwarding element, to implement a first logical port of a
logical router of the particular logical network, the first logical
port used only for egress traffic directed outside of the
managed network, the first managed forwarding element
implementing the first logical port by connecting directly to a
physical network element outside of the managed network in
order to send egress traffic directly to the physical network
element without the egress traffic passing through any
intervening managed forwarding elements in the managed
network; and
configuring a second managed forwarding element in the
managed network to implement a second logical port of the
2
Appeal 2017-008134
Application 14/068,658
logical router, the second logical port used for both ingress
traffic received from outside the managed network and egress
traffic directed outside the managed network, wherein the
second managed forwarding element receives ingress traffic
addressed to the first virtual machine directly from the physical
network element and transmits said ingress traffic to the first
managed forwarding element.
The Examiner relies on the following as evidence of unpatentability:
McDysan
Sanden
Nilakantan
Sridharan
Kannan
US 2010/0175125 A1
US 2012/0173757 A1
US 2012/0278802 A1
US 2013/0058346 A1
US 2013/0145002 A1
July 8, 2010
July 5, 2012
Nov. 1,2012
Mar. 7,2013
June 6, 2013
Martin Casado et al., Virtualizing the Network Forwarding Plane 1—6 (Dec.
2010) (“Casado”).3
The Rejections
Claims 1, 3, and 4 are rejected under 35 U.S.C. § 103(a) (pre-AIA) or
§ 103 as unpatentable over Nilakantan and Casado. Final Act. 3—10.
Claim 5 is rejected under 35 U.S.C. § 103(a) (pre-AIA) or § 103 as
unpatentable over Nilakantan, Casado, and McDysan. Final Act. 10-11.
Claims 6—11, 13—16, and 20—22 are rejected under 35 U.S.C. § 103(a)
(pre-AIA) or § 103 as unpatentable over Sridharan and Casado. Final Act.
11-23.
Claim 12 is rejected under 35 U.S.C. § 103(a) (pre-AIA) or § 103 as
unpatentable over Sridharan, Casado, and Sanden. Final Act. 24—25.
3 See also 6th Inf 1 CoNEXT 2010 Conference, ACM PRESTO Workshop
(Nov. 30, 2010), available at https://conferences.sigcomm.org/co-
next/2010/Workshops/PRESTO/PRESTO_papers/06-Casado.pdf.
3
Appeal 2017-008134
Application 14/068,658
Claim 19 is rejected under 35 U.S.C. § 103(a) (pre-AIA) or § 103 as
unpatentable over Sridharan, Casado, and Kannan. Final Act. 25—26.
OBVIOUSNESS REJECTION OVER NILAKANTAN AND CASADO
Regarding independent claim 1, the Examiner finds that Nilakantan
and Casado teach its limitations. In particular, the Examiner determines
Nilakantan teaches configuring a first managed forwarding element (MFE)
in a network (e.g., host 208 in Figs. 3—4 or 206 in Figs. 5—6). Final Act. 4
(citing Nilakantan, Figs. 3—6). The Examiner states Nilakantan does not
teach implementing a first logical port used only for egress traffic directed
outside the managed network and configuring a second MFE having a
second logical port used for both ingress traffic received from outside the
managed network and egress traffic directed outside the managed network.
Id. at 5. The Examiner turns to Casado, when combined with Nilakantan, to
teach these features. Id. at 5—8 (citing Casado, pp. 2—5, Fig. 1).
Among other arguments, Appellants argue Nilakantan and Casado do
not teach or suggest a method that configures a first MFE to implement a
first logical port of a logical router used to send only egress traffic outside of
the managed network. App. Br. 11—13 (citing Casado, pp. 2—4). Appellants
contend Casado does not mention logical ports restricted by ingress or
egress. Id. at 12—13. Appellants specifically assert Casado discusses
generic logical ports and does not differentiate between these discussed
ports, such that Casado fails “adequately [to] address the specifically recited
features of the first and second logical ports.” Id. at 13 (underlining
omitted).
4
Appeal 2017-008134
Application 14/068,658
ISSUE
Under § 103, has the Examiner erred in rejecting claim 1 by finding
that Nilakantan and Casado collectively would have taught or suggested a
method that configures (a) “a first managed forwarding element in the
network ... to implement a first logical port of a logical router . . . the first
logical port used only for egress traffic directed outside of the managed
network,” and (b) “a second managed forwarding element in the managed
network to implement a second logical port of the logical router, the second
logical port used for both ingress traffic received from outside the managed
network and egress traffic directed outside the managed network”?
ANALYSIS
Based on the record before us, we find error in the Examiner’s
rejection of claim 1. Regarding the recitation of “the first logical port used
only for egress traffic,” the Examiner cites both Nilakantan and Casado.
Ans. 2—3 (citing Nilakantan || 6, 30—31, Figs. 3—6 and Casado, p. 3).
Nilakantan teaches a host (e.g., 208) having a hypervisor (e.g., 302),
including a virtual machine (e.g., 300) and a virtual switch (e.g., 304), that is
mapped to the recited “first managed forwarding element.” Final Act. 4
(discussing, e.g., host 208 in Figure 3). Specifically, Nilakantan teaches
configuring a host’s network connections/setup, including configuring the
hypervisor’s switch (e.g., a part of the mapped MFE, discussed in Final Act.
4), by analyzing the network connectivity requirements of persona 252 and
creating virtual ports (e.g., logical ports) to access the VLANs. See
Nilakantan || 30—31, Figs. 2—3.
5
Appeal 2017-008134
Application 14/068,658
This suggests to one skilled in the art that switches can be configured
to transmit traffic in a given direction (e.g., egress traffic to access the
VLANs) based on a persona’s network connectivity requirements. See id.
However, Nilakantan does not provide a sufficient suggestion to include a
second managed forwarding element or the recited first and second ports as
noted by the Examiner. Final Act. 5. Nor does this citation adequately
suggest including two ports, one used only for egress traffic directed outside
of a managed network and a second used for both ingress and egress traffic
to and from the managed network as recited in claim 1. App. Br. 28 (Claims
App’x).
Casado also teaches a “network hypervisor” in the context of a logical
forwarding plane. Casado 2. Specifically, Casado teaches a network
hypervisor maps a logical forwarding plane into physical hardware, and such
logical forwarding plane is the network’s logical abstraction (e.g., lookup
tables, ports, and primitives for forwarding packets). Id. at 2—3. For
example, Casado discusses forwarding elements containing a set of logical
ports and discusses both “ingress ports” and “egress ports.” Id. at 3^4
(stating both (i) “[i]t could be . . . the ingress port” and (ii) “the network
must send the packets to the physical objects to which these egress ports are
bound”); see also id. at 4 (further discussing “the logical egress port” and
“table to map a logical egress port to a physical location”). As such, Casado
teaches that a hypervisor, such as Nilakantan’s hypervisor 302, can include a
forwarding element having (1) an egress port that sends traffic to a physical
location or (2) an ingress port. See id. at 3^4.
Even presuming that Casado’s egress port teaches or suggests a port
restricted to egress or the recited “first logical port used only for egress
6
Appeal 2017-008134
Application 14/068,658
traffic,” Casado does not further teach or suggest a second logical port on a
second managed MFE used for both ingress traffic and egress traffic as
additionally recited in claim 1. See App. Br. 13. At best, Casado teaches
ports may appear and leave dynamically and can be ingress or egress ports.
Casado 3, cited in Final Act. 6. We therefore agree with Appellants that
Casado discusses “generic logical ports” and “does not adequately address
the specifically recited features of the first and second logical ports.” App.
Br. 13 (underlining omitted); see also Reply Br. 3. That is, Casado does not
discuss a logical port that provides both ingress and egress traffic. Casado
3^4. Nor does the Examiner sufficiently elaborate on this point. See Ans.
2—6 (providing no response related to “the second logical port used for both
ingress traffic received from outside the managed network and egress traffic
directed outside of the managed system” as recited).
Granted, Casado teaches a table to map incoming packets to their
logical forwarding context. Casado 4, cited in both Final Act. 7 and Ans. 5.
Nonetheless, the Examiner has not explained sufficiently how one skilled in
the art would have recognized this teaching of mapping an incoming packet
to logical context suggests the recited “second logical port used for . . .
ingress traffic received” along with “egress traffic directed outside the
managed network” in claim 1. Moreover, even presuming without agreeing
that Casado’s general teaching of a table that maps logical egress ports to
physical locations suggests the recited “first logical port... to a physical
network element outside of the managed network” as the Examiner
determines (Ans. 6),4 there is an insufficient explanation in the record that
4 Nilakantan teaches and shows in Figure 3 hypervisor 302 within host 208
is connected to VLANs 240, 242. These VLANs are described as part of
7
Appeal 2017-008134
Application 14/068,658
this mapping of an incoming packet to logical context also teaches or
suggests the recited “second logical port. . . used for . . . ingress traffic
received from outside the managed network” in claim 1.
For the foregoing reasons, Appellants have persuaded us of error in
the rejection of (1) independent claim 1 and (2) dependent claims 3 and 4 for
similar reasons.
OBVIOUSNESS REJECTION OVER NILAKANTAN, CAS ADO,
AND MCDYSAN
Claim 5 depends from claim 1 and is rejected under 35 U.S.C.
§ 103(a) (pre-AIA) or § 103 as unpatentable over Nilakantan, Casado, and
McDysan. Final Act. 10—11. The rejection does not rely on McDysan to
teach or suggest the above-noted deficiencies of the Nilakantan-Casado
rejection. See id. As such, for reasons similar to those above for claim 1,
we will not sustain the rejection of claim 5.
OBVIOUSNESS REJECTION OVER SRIDHARAN AND CASADO
Claims 6—11
Independent claim 6 recites “a second managed forwarding element
. . . performed logical processing on the first packet as having received the
first packet at a first logical port of a logical router, the first logical port used
for both ingress traffic received from outside the managed network and
egress traffic directed outside the managed network” and “performing
managed system 200 in Figure 2. Nilakantan 122, Figs. 2—3. Casado
discusses using the hypervisor within a multi-tenant network environment,
describing the tenant networks separately from the VLANs. Casado 5.
8
Appeal 2017-008134
Application 14/068,658
logical processing on the second packet to logically send the packet to a
second logical port of the logical router that is used only for egress traffic
directed outside of the managed network.” App. Br. 29 (Claims App’x).
These recitations are similar to those discussed above for claim 1.
For this claim, the Examiner relies on Sridharan to teach the recited
program having instructions that (1) when executed by a processing unit,
implements a first MFE (e.g., NVGRE (network virtualization Generic
Routing Encapsulation) router shown in Figure 5 as 509, 510 and Figure 10
at 1041—1044, discussed in Final Act. 11) and (2) receives a first packet
from a second MFE from a particular source through a physical network
outside the managed network (NVGRE routers 1041—1044 between host
1003—1005 and devices on network 1002, discussed in Final Act. 12). The
Examiner further states Sridharan does not disclose the first or second
logical ports or that the second logical port is “used only for egress traffic
directed outside of the managed network” as recited, turning to Casado.
Final Act. 13—14 (citing Casado 3—4).
Appellants present arguments similar to claim 1. Among them,
Appellants argue Casado only provides a general teaching related to logical
ports and fails to address the specific features of the recited first and second
logical ports. App. Br. 16—17.
ISSUE
Under § 103, has the Examiner erred in rejecting claim 6 by finding
that Sridharan and Casado teach or suggest a
medium storing a program which when executed by at least one
processing unit of a host machine implements a first managed
forwarding element. . . the program comprising sets of
9
Appeal 2017-008134
Application 14/068,658
instructions for: receiving a first packet from a second managed
forwarding element, wherein the second managed forwarding
element. . . performed logical processing on the first packet as
having received the first packet at a first logical port of a logical
router, the first logical port used for both ingress traffic
received from outside the managed network and egress traffic
directed outside the managed network . . . [and] performing
logical processing on the second packet to logically send the
packet to a second logical port of the logical router that is used
only for egress traffic directed outside of the managed
network[?]
ANALYSIS
Although the Examiner cites Sridharan—not Nilakantan—to teach
some of the features of claim 6, the rejection still relies on Casado to teach
the recitations related to the “first” and “second logical” ports. Final Act.
11—14; see also Ans. 7—8. As noted above, Casado’s general teaching of
logical ports (e.g., ingress or egress ports) does not sufficiently teach or
suggest to one skilled in the art the recited “first logical port used for both
ingress traffic received from outside the managed network and egress traffic
directed outside the managed network” in claim 6. We refer to the above
analysis for details.
Also, to the extent Sridharan teaches transmitting packets from a MFE
to “a physical network element outside of the managed network” as recited
(Final Act. 12 (citing Sridharan | 68 and discussing hosts 1003—1005 and
corporate network 1002)), we agree this teaching does not sufficiently
demonstrate transmitting packets to the physical network element using a
connection “between the host machine and the physical network element
that does not include any intervening managed forwarding elements” as
recited. App. Br. 17—18. “In fact, the cited Figure 10 illustrates that a
10
Appeal 2017-008134
Application 14/068,658
virtual network gateway” having a NVGRE router “is located between” a
host machine (e.g., 1003) and the physical network element (e.g., 1002). Id.
at 18 (discussing Sridharan | 68, Fig. 10). To be sure, Sridharan also
teaches delivering packets between hosts “without going through a separate
virtual router on the datacenter network.'1'’ Sridharan 148 (emphasis added).
Yet, the record does not clearly explain how paragraph 48’s teaching applies
to sending packets to a destination host located outside the datacenter
network, such as devices (e.g., computer 1 or Cl) on Figure 10’s corporate
network 1002. Compare id. with id. 63—68, Fig. 10. As such, based on
the record, Sridharan does not teach or suggest “transmitting the second
packet directly from the” MFE “to the physical network element via a
connection . . . between the host machine and the physical network element
that does not include any intervening managed forwarding elements” in
claim 6.
For the foregoing reasons, Appellants have persuaded us of error in
the rejection of (1) independent claim 6 and (2) dependent claims 7—11 for
similar reasons.
Claims 13—16 and 20—22
Independent claim 13 recites a method with similar recitations to
claim 6 and is rejected similarly based on Sridharan and Casado. Final Act.
14—18. Appellants present analogous arguments for claim 13 as claim 6.
App. Br. 18—22. We are persuaded of error for similar reasons to those
previously discussed. Accordingly, Appellants have persuaded us of error in
the rejection of (1) independent claim 13 and (2) dependent claims 14—16
and 20-22.
11
Appeal 2017-008134
Application 14/068,658
THE REMAINING OBVIOUSNESS REJECTIONS
Claim 12 indirectly depends from claim 6 and is rejected under 35
U.S.C. § 103(a) (pre-AIA) or § 103 as unpatentable over Sridharan, Casado,
and Sanden (Final Act. 24—25); claim 19 depends from claim 13 and is
rejected under 35 U.S.C. § 103(a) (pre-AIA) or § 103 as unpatentable over
Sridharan, Casado, and Kannan. Id. at 25—26. These rejections do not rely
on Sanden or Kannan to illustrate the above-noted deficiencies of the
Sridharan-Casado rejection. See id. at 24—26. As such, for reasons similar
to those above for claims 6 and 13, we will not sustain the rejections of
claims 12 and 19.
DECISION
We reverse the Examiner’s rejections of claims 1, 3—16, and 19—22
under § 103.
Pending claims 17 and 18 have been objected to but would be
allowable if rewritten in independent form to include all of the limitations of
their base claim and any intervening claims.
REVERSED
12