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 .
Detail Action
This office action is in response to the application filed on 4/15/2024.
Preliminary amendment filed on 8/16/2024 is acknowledged.
Claims 1-20 are pending.
Claim Rejections - 35 USC § 103
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, 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.
Claim(s) 1-8, 10-17 and 19-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chakra (“Advancing Performance Benchmarking and Co-design using Standardized Execution Trace“, per IDS) in view of ENRICI (US 2022/0345535 Al).
Per claim 1,
Chakra discloses
receiving, at a test platform, a graph-based representation of AI/ML workload execution comprising a collective communication node; (pp. 5, see ML framework collecting execution traces of ML model…identify communication nodes in PyTorch, “3.3.3” see execution graphs, “3.3.1-2” discloses multi-nodes execution )
implementing, by the test platform, graph-based representation of AI/ML workload in an emulated test case using an emulation engine. (section 3.3 discloses simulating and emulating using execution graph. Chakra section “5.4 Test Case generator” for generating test case)
Chakra does not, however, ENRICI discloses
expanding, by the test platform, the collective communication node by replacing a collective communication operation of the collective communication node with low-level processing instructions; ([0080], disclose partition data-flow graph G of a ML application into sets of nodes and edges…send/receive primitives…create a graph G’ that is reducible, where G’ is modified from G and it is reducible corresponding to expanded; see also [0081-0099])
generating, by the test platform, a modified graph-based representation of AI/ML workload execution comprising the low-level processing instructions(continue, graph G’ corresponds to modified graph comprising low-level instructions)
implementing, modified graph-based representation of AI/ML workload (see above)
Therefore, it would have been obvious to a person of ordinary skill before the effective filing date of the invention to incorporate the teachings of ENRICI into the teachings of Chakra to include the limitation disclosed by ENRICI. The modification would be obvious to one of ordinary skill in the art to want to evaluate performance of scheduling solutions as suggested by ENRICI ([0114])
Per claim 2, the rejection of claim 1 is incorporated.
Chakra/ ENRICI discloses
wherein the low-level processing instructions comprise send and receive primitives. (ENRICI, [0080], send/receive primitives)
Per claim 3, the rejection of claim 2 is incorporated.
Chakra/ ENRICI discloses
wherein expanding the collective communication node comprises replacing the collective communication node with send and receive nodes. (ENRICI, [0083-84], see transform graph G into graph G’ with new nodes and edges)
Per claim 4, the rejection of claim 3 is incorporated.
Chakra/ ENRICI discloses
displaying a representation of the expanded collective communication node for a single rank. (ENRICI , [0080], disclose partitioned graph G’ where nodes are subgraphs of graph G. [0114], see single partitioned graph G’. [0121] see user output devic
Per claim 5, the rejection of claim 1 is incorporated.
Chakra/ ENRICI discloses
comprising defining a collective communication algorithm based on the low-level processing instructions, wherein the emulated test case uses the collective communication algorithm. (ENRICI, [0079-0080], discloses partitioning attempts to match the same objective function and therefore the same communication algorithm.)
Per claim 6, the rejection of claim 1 is incorporated.
Chakra/ ENRICI discloses
reporting at least one performance metric from the executed emulated test case. (Chakra, see “6 USE CASES…benchmark to generate…performance and system-level metrics)
Per claim 7, the rejection of claim 1 is incorporated.
Chakra/ ENRICI discloses
wherein the low-level processing instructions are based on the collective communication operation. (ENRICI , [0079-0080], see objective function corresponding to collective communication operation partitioned into send/receive primitives.)
Per claim 8, the rejection of claim 7 is incorporated.
Chakra/ ENRICI discloses
comprising revising the low-level processing instructions to define a revised collective communication algorithm. (ENRICI, [0080], discloses selecting a different algorithm corresponding to revision of via repeating)
Per claims 10-17, see rejections of claims 1-8.
Per claims 19-20, see rejections of claims 1 and 3.
Claim(s) 9 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Chakra (“Advancing Performance Benchmarking and Co-design using Standardized Execution Trace“, per IDS) in view of ENRICI (US 2022/0345535 Al) and further in view of Tellado et al. (US 20180249406 A1)
Per claim 9, the rejection of claim 8 is incorporated.
Chakra/ ENRICI discloses
a first executed emulated test case using the low-level processing instructions based on the collective communication operation and from a second executed emulated test case using the revised collective low-level processing instructions. ([0080] discloses different algorithm corresponding a first and a second emulated test cases.)
Chakra/ ENRICI does not, however, Tellado discloses
comparing performance metrics of the first and second test cases.( [0033], disclose performance metrics of selected test cases are compared. [0065] so as to select the best metrics.)
Therefore, it would have been obvious to a person of ordinary skill before the effective filing date of the invention to incorporate the teachings of Tellado into the teachings of Chakra/ ENRICI to include the limitation disclosed by Tellado. The modification would be obvious to one of ordinary skill in the art to want to select the best metrics as suggested by Tellado ([0065])
Per claim 18, see rejection of claim 9.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Philip Wang whose telephone number is 571-272-5934. The examiner can normally be reached on Monday – Friday 8:00AM -4:00PM. Any inquiry of general nature or relating to the status of this application should be directed to the TC2100 Group receptionist: 571-272-2100.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Lewis Bullock, can be reached at 571-272-3759. 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 the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/PHILIP WANG/Primary Examiner, Art Unit 2199
Prosecution Insights