DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Applicant’s claim for the benefit of a prior-filed application under 35 U.S.C. 119(e) or under 35 U.S.C. 120, 121, 365(c), or 386(c) is acknowledged.
Information Disclosure Statement
3. The information disclosure statement(s) submitted on May 20, 2024 has been considered by the Examiner and made of record in the application file.
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
Claim Rejections - 35 USC § 103
4. The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102 of this title, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
The factual inquiries set forth in Graham v. John Deere Co., 383 U.S. 1, 148 USPQ 459 (1966), that are applied for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claims 1-3 and 8 are rejected under 35 U.S.C. 103 as being unpatentable over Parker et al. (U.S. Patent Application Publication # 2016/0294694 A1) in view of Roweth et al. (U.S. Patent Application Publication # 2022/0329521 A1).
Regarding claim 1, Parker et al. teach a network (read as dragonfly network (Paragraph [0133])) comprising:
a plurality of local endpoints (Fig(s).1, 3-4, 9, and 12) in a local group (read as local group (Paragraph [0149]));
a plurality of remote endpoints (Fig(s).1, 3-4, 9, and 12) in a remote group (read as a remote group (Paragraph [0135]));
a switch comprising a port (read as port (Fig(s).4, 9 and 12)) for a link between the local group and the remote group (read as a subswitch (Fig(s).4, 9 and 12; Paragraph [0108]) Also, Parker et al. teach “a variety of tables to determine paths available in routing a packet or message, and provide routing flexibility in the dragonfly network configuration. Different tables exist to provide routing within a group and between groups, and for minimal and non-minimal routing paths.”(Paragraph [0133]) For example, “This table should only contain routes to other router chips or optical port numbers that ultimately lead to an intermediate group that can safely route to all other groups in the system. The table also provides the mechanism that distributes non-minimal traffic roughly evenly over the groups in the system.”(Paragraph [0138])); and
However, Parker et al. fail to explicitly teach logic to:
convert traffic received from the local endpoints to a bandwidth demand for one or more destination endpoint in a remote group; and
determine a sum over the one or more destination endpoints of a minimum of (a) a maximum bandwidth of the link, and (b) a bandwidth demand to one or more of the remote endpoints.
Roweth et al. teach a switch (Fig.2 @ 202) comprising of logic to:
convert traffic received from the local endpoints to a bandwidth demand for one or more destination endpoint in a remote group (read as “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])); and
determine a sum over the one or more destination endpoints of a minimum of (a) a maximum bandwidth of the link (read as maximum bandwidth (Paragraph [0035])), and (b) a bandwidth demand (read as bandwidth allocation (Paragraph [0035])) to one or more of the remote endpoints. (read as “If a class does not use its minimum bandwidth, other classes may use the unused bandwidth, but no class can get more than its maximum allocated bandwidth. The ability to manage bandwidth provides the opportunity to dedicate network resources, as well as CPUs and memory bandwidth to a particular application.”(Paragraph [0035]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Regarding claim 2, and as applied to claim 1 above, Parker et al., as modified by Roweth et al., teach a network (Fig(s).1, 3-4, 9, and 12) further comprising:
logic to determine a portion of the traffic received from the local endpoints to route non-minimally to one or more of the remote endpoints. (read as routing algorithm for generating non-minimal routing table(s) (Paragraph(s) [0065], [0133] and [0149]))
Regarding claim 3, and as applied to claim 2 above, Parker et al., as modified by Roweth et al., teach a network (Fig(s).1, 3-4, 9, and 12) wherein the portion of the traffic to route non-minimally is determined from a ratio of the maximum bandwidth of the link and the sum. (read as “Routing choices are presented via tables in one example, and may be biased toward certain routes or toward minimal or non-minimal routes depending on the network configuration and state.”(Paragraph [0105]))
Regarding claim 8, and as applied to claim 1 above, Parker et al. teach a network (Fig(s).1, 3-4, 9, and 12) further comprising:
a plurality of remote endpoint groups (read as a remote group (Paragraph [0135])); and
However, Parker et al. fail to explicitly teach logic to:
convert packet flow rates and packet trajectories for the port into aggregate contention metrics for the link; and
broadcast the contention metrics to the remote endpoint groups.
Roweth et al. teach a network switch (Fig.2 @ 202) with logic to: convert packet flow rates and packet trajectories for the global link into aggregate contention metrics for the global link (Fig.2 @ 204); and
broadcast the contention metrics to the one or more destination endpoints. (Fig.2 @ 208, 220)
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for bandwidth allocation based on a maximum bandwidth allocation limit and switch computer architecture as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Claims 4-7 are rejected under 35 U.S.C. 103 as being unpatentable over Parker et al. (U.S. Patent Application Publication # 2016/0294694 A1), in view of, and Roweth et al. (U.S. Patent Application Publication # 2022/0329521 A1), and Birrittella et al. (U.S. Patent Application Publication # 2015/0222533 A1).
Regarding claim 4, and as applied to claim 1 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
However, Parker et al. and Roweth et al. fail to explicitly teach logic to determine a rate of non-minimally routed traffic on the link.
Birrittella et al. teach logic to determine a rate of non-minimally routed traffic on the link. (read as packet data rate matching (Paragraph [0446]) Also, “When optimizing for bandwidth, non-minimal routing may be utilized for spreading the traffic to reduce congestion.”(Paragraph [0347]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for packet data rate matching as taught by Birrittella et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Regarding claim 5, and as applied to claim 4 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
However, Parker et al. and Roweth et al. fail to explicitly teach wherein the rate of non-minimally routed traffic is endpoint independent.
Birrittella et al. teach a method wherein the rate of non-minimally routed traffic is endpoint independent. (read as packet data rate matching (Paragraph [0446]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for packet data rate matching as taught by Birrittella et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Claims 6-7 are rejected under 35 U.S.C. 103 as being unpatentable over Parker et al. (U.S. Patent Application Publication # 2016/0294694 A1), in view of Roweth et al. (U.S. Patent Application Publication # 2022/0329521 A1), Birrittella et al. (U.S. Patent Application Publication # 2015/0222533 A1), and Abts et al. (U.S. Patent Application Publication # 2023/0161621 A1).
Regarding claim 6, and as applied to claim 4 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
Birrittella et al. teach “Method, apparatus, and systems for reliably transferring Ethernet packet data over a link layer and facilitating fabric-to-Ethernet and Ethernet-to-fabric gateway operations at matching wire speed and packet data rate.”(Abstract)
However, Parker et al., Roweth et al, and Birrittella et al. fail to explicitly teach logic to determine an injection rate throttle setting for the non-minimally routed traffic based on the rate of non-minimally routed traffic on the link.
Abts et al. teach logic to determine an injection rate throttle setting for the non-minimally routed traffic based on the rate of non-minimally routed traffic on the link. (read as “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]) For example, “Where an amount of data to be transferred exceeds the capacity for a given link, the compiler, having complete knowledge of the system topology as well as a complete view of the state of the network during each cycle, may select one or more non-minimal routes in addition to or instead of a direct route through which to route data from the first TSP to the second TSP. ”(Paragraph [0080]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al., the function for packet data rate matching as taught by Birrittella et al., and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Regarding claim 7, and as applied to claim 6 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
Birrittella et al. teach “Method, apparatus, and systems for reliably transferring Ethernet packet data over a link layer and facilitating fabric-to-Ethernet and Ethernet-to-fabric gateway operations at matching wire speed and packet data rate.”(Abstract)
However, Parker et al., Roweth et al, and Birrittella et al. fail to explicitly teach wherein the endpoint throttle setting is determined from a ratio of the maximum bandwidth of the link and the rate of non-minimally routed traffic on the link.
Abts et al. teach a method wherein the endpoint throttle setting is determined from a ratio of the maximum bandwidth of the link and the rate of non-minimally routed traffic on the link. (read as “the system further comprises a compiler configured to explicitly schedule communication traffic across the global and local links of the network of processors, with explicit send or receive instructions executed at specific times to establish a specific ordering of operations performed by the network of processors.”(Paragraph [0004]) For example, “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al., the function for packet data rate matching as taught by Birrittella et al., and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Claims 9-12 and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Parker et al. (U.S. Patent Application Publication # 2016/0294694 A1), in view of Roweth et al. (U.S. Patent Application Publication # 2022/0329521 A1), and Abts et al. (U.S. Patent Application Publication # 2023/0161621 A1).
Regarding claim 9, and as applied to claim 1 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
However, Parker et al. and Roweth et al. fail to explicitly teach logic to determine an injection rate throttle setting for the local endpoints that reduces the depth of a queue for the port.
Abts et al. teach logic to determine an injection rate throttle setting for the local endpoints that reduces the depth of a queue for the port. (read as “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]) For example, “Where an amount of data to be transferred exceeds the capacity for a given link, the compiler, having complete knowledge of the system topology as well as a complete view of the state of the network during each cycle, may select one or more non-minimal routes in addition to or instead of a direct route through which to route data from the first TSP to the second TSP. ”(Paragraph [0080]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Regarding claim 10, Parker et al. teach a network switch (Fig(s).3, 4, 9, and 12) comprising:
a plurality of local ports (Fig(s).3-4, 9, and 12) to endpoints of a local group (read as local group (Paragraph [0149]));
a global port (Fig(s).3-4, 9, and 12) for a global link (read as global channel(s) (Fig.3-4, 7, 9, and 12));
However, Parker et al. fail to explicitly teach logic to:
convert traffic received from the endpoints of the local group to a bandwidth demand for one or more destination endpoints in a remote group; and
on condition that the bandwidth demand indicates a contention condition,
signal one of more of the endpoints of the local group to perform one or both of non-minimal routing and injection throttling.
Roweth et al teach a switch executing logic to:
convert traffic received from the endpoints of the local group to a bandwidth demand for one or more destination endpoints in a remote group (read as “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])); and
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
However, Parker et al. and Roweth et al. fail to explicitly teach on condition that the bandwidth demand indicates a contention condition,
signal one of more of the endpoints of the local group to perform one or both of non-minimal routing and injection throttling.
Abts et al teach network processor(s) executing logic to:
on condition that the bandwidth demand indicates a contention condition (read as “the amount of bandwidth injected is no longer limited by the injection bandwidth, but by the switch router bandwidth, which also impacts non-minimal routing.”(Fig.3C; Paragraph [0082])),
signal one of more of the endpoints of the local group to perform one or both of non-minimal routing and injection throttling. (read as “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]) For example, “Where an amount of data to be transferred exceeds the capacity for a given link, the compiler, having complete knowledge of the system topology as well as a complete view of the state of the network during each cycle, may select one or more non-minimal routes in addition to or instead of a direct route through which to route data from the first TSP to the second TSP. ”(Paragraph [0080]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Regarding claim 19, Parker et al. teach a network contention control process in a Dragonfly network (Fig(s).3, 4, 9, and 12), the process comprising:
in a switch of the Dragonfly network (read as subswitch (Fig(s).3, 4, 9, and 12)),
However, Parker et al. fail to explicitly teach converting traffic received from endpoints of a local group of the switch to a bandwidth demand for one or more destination endpoints in a remote group; and
on condition that the bandwidth demand indicates a contention condition,
generating from the switch a signal one of more of the endpoints of the local group to perform one or both of non-minimal routing and injection throttling.
Roweth et al teach a switch for converting traffic received from the endpoints of the local group to a bandwidth demand for one or more destination endpoints in a remote group (read as “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])); and
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
However, Parker et al. and Roweth et al. fail to explicitly teach on condition that the bandwidth demand indicates a contention condition,
generating from the switch a signal one of more of the endpoints of the local group to perform one or both of non-minimal routing and injection throttling.
Abts et al teach network processor(s) executing logic to:
on condition that the bandwidth demand indicates a contention condition (read as “the amount of bandwidth injected is no longer limited by the injection bandwidth, but by the switch router bandwidth, which also impacts non-minimal routing.”(Fig.3C; Paragraph [0082])),
generating from the switch a signal one of more of the endpoints of the local group to perform one or both of non-minimal routing and injection throttling. (read as “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]) For example, “Where an amount of data to be transferred exceeds the capacity for a given link, the compiler, having complete knowledge of the system topology as well as a complete view of the state of the network during each cycle, may select one or more non-minimal routes in addition to or instead of a direct route through which to route data from the first TSP to the second TSP. ”(Paragraph [0080]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Regarding claims 11 and 20, and as applied to claims 10 and 19 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Abts et al. teach “the system further comprises a compiler configured to explicitly schedule communication traffic across the global and local links of the network of processors, with explicit send or receive instructions executed at specific times to establish a specific ordering of operations performed by the network of processors.”(Paragraph [0004]) Also, Abts et al. teach “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079])
However, Parker et al. and Abts et al. fail to explicitly teach wherein the contention condition is determined as a sum over the one or more destination endpoints of a minimum of (a) a maximum bandwidth of the global link, and (b) a bandwidth demand to one or more of the destination endpoints.
Roweth et al. teach a method wherein the contention condition is determined as a sum over the one or more destination endpoints of a minimum of (a) a maximum bandwidth of a global link to the remote group (read as maximum bandwidth (Paragraph [0035])), and (b) a bandwidth demand to one or more of the destination endpoints.(read as “If a class does not use its minimum bandwidth, other classes may use the unused bandwidth, but no class can get more than its maximum allocated bandwidth. The ability to manage bandwidth provides the opportunity to dedicate network resources, as well as CPUs and memory bandwidth to a particular application.”(Paragraph [0035]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Regarding claim 12, and as applied to claim 10 above, Parker et al., as modified by Roweth et al. and Abts et al., teach a network switch (Fig(s).1, 3-4, 9, and 12) further comprising:
logic to determine and communicate to the endpoints of the local group an extent of traffic to route non-minimally to the one or more of the destination endpoints. (read as routing algorithm for generating non-minimal routing table(s) (Paragraph(s) [0065], [0133] and [0149]))
Regarding claim 13, and as applied to claim 12 above, Parker et al., as modified by Roweth et al. and Abts et al., teach a network switch (Fig(s).1, 3-4, 9, and 12) wherein the extent of the traffic to route non-minimally is determined from a ratio of the maximum bandwidth of the global link and the sum. (read as “Routing choices are presented via tables in one example, and may be biased toward certain routes or toward minimal or non-minimal routes depending on the network configuration and state.”(Paragraph [0105]))
Regarding claim 18, and as applied to claim 10 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Abts et al. teach “the system further comprises a compiler configured to explicitly schedule communication traffic across the global and local links of the network of processors, with explicit send or receive instructions executed at specific times to establish a specific ordering of operations performed by the network of processors.”(Paragraph [0004]) Also, Abts et al. teach “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079])
However, Parker et al. and Abts et al. fail to explicitly teach logic to:
convert packet flow rates and packet trajectories for the global link into aggregate contention metrics for the global link; and
broadcast the contention metrics to the one or more destination endpoints.
Roweth et al. teach a network switch (Fig.2 @ 202) further comprising:
logic to: convert packet flow rates and packet trajectories for the global link into aggregate contention metrics for the global link (Fig.2 @ 204); and
broadcast the contention metrics to the one or more destination endpoints. (Fig.2 @ 208, 220)
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al. and the function for bandwidth allocation based on a maximum bandwidth allocation limit and switch computer architecture as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Claims 14-17 are rejected under 35 U.S.C. 103 as being unpatentable over Parker et al. (U.S. Patent Application Publication # 2016/0294694 A1), in view of Roweth et al. (U.S. Patent Application Publication # 2022/0329521 A1), Abts et al. (U.S. Patent Application Publication # 2023/0161621 A1), and Birrittella et al. (U.S. Patent Application Publication # 2015/0222533 A1).
Regarding claim 14, and as applied to claim 10 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
Abts et al. teach “the system further comprises a compiler configured to explicitly schedule communication traffic across the global and local links of the network of processors, with explicit send or receive instructions executed at specific times to establish a specific ordering of operations performed by the network of processors.”(Paragraph [0004]) Also, Abts et al. teach “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079])
However, Parker et al., Roweth et al., and Abts et al. fail to explicitly teach logic to determine a rate of non-minimally routed traffic on the global link.
Birrittella et al. teach logic to determine a rate of non-minimally routed traffic on the global link. (read as packet data rate matching (Paragraph [0446]) Also, “When optimizing for bandwidth, non-minimal routing may be utilized for spreading the traffic to reduce congestion.”(Paragraph [0347]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for packet data rate matching as taught by Birrittella et al., the function for non-minimal routing as taught by Abts et al., and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Regarding claim 15, and as applied to claim 14 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
Abts et al. teach “the system further comprises a compiler configured to explicitly schedule communication traffic across the global and local links of the network of processors, with explicit send or receive instructions executed at specific times to establish a specific ordering of operations performed by the network of processors.”(Paragraph [0004]) Also, Abts et al. teach “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079])
However, Parker et al., Roweth et al., and Abts et al. fail to explicitly wherein the rate of non-minimally routed traffic is endpoint-independent.
wherein the rate of non-minimally routed traffic is endpoint-independent. (read as packet data rate matching (Paragraph [0446]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for packet data rate matching as taught by Birrittella et al., the function for non-minimal routing as taught by Abts et al., and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of enhancing transmission resource allocation by devices in a Dragonfly network.
Regarding claim 16, and as applied to claim 14 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
Birrittella et al. teach “Method, apparatus, and systems for reliably transferring Ethernet packet data over a link layer and facilitating fabric-to-Ethernet and Ethernet-to-fabric gateway operations at matching wire speed and packet data rate.”(Abstract)
However, Parker et al., Roweth et al, and Birrittella et al. fail to explicitly teach logic to determine an injection traffic throttle setting at the endpoints of the local group for the non-minimally routed traffic based on the rate of non-minimally routed traffic on the global link.
Abts et al. teach logic to determine an injection traffic throttle setting at the endpoints of the local group for the non-minimally routed traffic based on the rate of non-minimally routed traffic on the global link. (read as “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]) For example, “Where an amount of data to be transferred exceeds the capacity for a given link, the compiler, having complete knowledge of the system topology as well as a complete view of the state of the network during each cycle, may select one or more non-minimal routes in addition to or instead of a direct route through which to route data from the first TSP to the second TSP. ”(Paragraph [0080]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al., the function for packet data rate matching as taught by Birrittella et al., and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Regarding claim 17, and as applied to claim 16 above, Parker et al. teach “A multiprocessor computer system comprises a dragonfly processor interconnect network that comprises a plurality of processor nodes and a plurality of routers. The routers are operable to route data by selecting from among a plurality of network paths from a target node to a destination node in the dragonfly network based on one or more routing tables.”(Fig(s).1, 3-4, 9, and 12; Abstract)
Roweth et al. teach “Switches can be configured in a hierarchical topology having a plurality of groups, where switches in a group are connected to one another, and groups are connected to other groups. Routing can be performed by maintaining per-group group load information. A packet can be routed between at least two groups using the per-group group load information to effect a set of routing decisions. The set of routing decisions can be biased towards or away one or more paths.” (Fig.2 @ 202; Abstract) For example, “Switch 202 can provide system-wide Quality of Service (QoS) classes, along with the ability to control how network bandwidth is allocated to different classes of traffic, and to different classes of applications, where a single privileged application may access more than one class of traffic.”(Fig.2 @ 202; Paragraph [0035])
Birrittella et al. teach “Method, apparatus, and systems for reliably transferring Ethernet packet data over a link layer and facilitating fabric-to-Ethernet and Ethernet-to-fabric gateway operations at matching wire speed and packet data rate.”(Abstract)
However, Parker et al., Roweth et al, and Birrittella et al. fail to explicitly teach wherein the endpoint throttle setting is determined from a ratio of a maximum bandwidth of the global link and the rate of non-minimally routed traffic on the global link.
Abts et al. teach a method wherein the endpoint throttle setting is determined from a ratio of a maximum bandwidth of the global link and the rate of non-minimally routed traffic on the global link. (read as “the system further comprises a compiler configured to explicitly schedule communication traffic across the global and local links of the network of processors, with explicit send or receive instructions executed at specific times to establish a specific ordering of operations performed by the network of processors.”(Paragraph [0004]) For example, “the multi-TSP system uses “non-minimal” routing that takes advantage of abundant path diversity exhibited by the Dragonfly-based hierarchical topology of the system, to spread the offered traffic across multiple injection links in each TSP.”(Paragraph [0079]))
Therefore, it would have been obvious to a person of ordinary skill in the art before the effective filing date of the claimed invention to employ the function for non-minimal routing as taught by Abts et al., the function for packet data rate matching as taught by Birrittella et al., and the function for bandwidth allocation based on a maximum bandwidth allocation limit as taught by Roweth et al. with the Dragonfly topology comprising of one or more subswitch(es) as taught by Parker et al. for the purpose of improving forwarding data traffic by devices in a Dragonfly network.
Conclusion
5. The prior art made of record and not relied upon is considered pertinent to Applicant’s disclosure:
D. Xiang, B. Li and Y. Fu ("Fault-Tolerant Adaptive Routing in Dragonfly Networks", 1 March-April 2019) teaches “A new deadlock free adaptive fault-tolerant routing algorithm based on a new two-layer safety information model, is proposed by mapping routers in a group, and groups of the dragonfly network into two separate hypercubes. The new fault-tolerant routing algorithm tolerates static and dynamic faults. Our method can determine whether a packet can reach the destination at the source by using the new safety information model, which avoids dead-ends and aimless misrouting.”(Fig(s).1-3; Abstract)
Maglione-Mathey, Yebenes, Escudero-Sahuquillo, Garcia, Quiles and Zahavi ("Scalable Deadlock-Free Deterministic Minimal-Path Routing Engine for InfiniBand-Based Dragonfly Networks", 1 Jan. 2018) teaches “a new deterministic, minimal-path routing for Dragonfly that prevents deadlocks using VLs according to the IB specification, so that it can be straightforwardly implemented in IB-based networks.”(Fig(s)2, 6, and 9; Abstract)
J. Escudero-Sahuquillo et al. ("Combining Congested-Flow Isolation and Injection Throttling in HPC Interconnection Networks", 2011) teaches “a method that combines injection throttling and congested-flow isolation.”(Fig(s).3-4; Abstract)
Any response to this Office Action should be faxed to (571) 273-8300 or mailed to:
Commissioner for Patents
P.O. Box 1450
Alexandria, VA 22313-1450
Any inquiry concerning this communication or early communications from the Examiner should be directed to Salvador E. Rivas whose telephone number is (571) 270-1784. The examiner can normally be reached on Monday-Friday from 7:00AM to 3:30PM.
If attempts to reach the Examiner by telephone are unsuccessful, the Examiner’s supervisor, Un C. Cho can be reached on (571) 272- 7919. The fax phone number for the organization where this application or proceeding is assigned is (571) 273-8300.
Information regarding the status of an application may be obtained from Patent Center. Status information for published applications may be obtained from Patent Center. Status information for unpublished applications is available through Patent Center to authorized users only. Should you have questions about access to the USPTO patent electronic filing system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
Any inquiry of a general nature or relating to the status of this application or proceeding should be directed to the receptionist/customer service whose telephone number is (571) 272-2600.
/SALVADOR E RIVAS/Primary Examiner, Art Unit 2413
June 26, 2026
Prosecution Insights