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
Application claims priority to foreign application with application number CN202410524402.5 dated 04/28/2024. Copies of certified papers required by 37 CFR 1.55 have been received. Priority is acknowledged under 35 USC 119(e) and 37 CFR 1.78.
Claim Objections
Claim 14 objected to because of the following informalities: the word "wherein" is repeated ("wherein: wherein the actual usage is a first actual usage..."). Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 13 and 15 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 13 appears to consist of two distinct claims, the second of which is not properly numbered and depends on claim 13 itself. Claim 15 is also dependent on itself, likely due to the previous error in numbering.
For purposes of examination, “Claim 13” will be used to refer to the first claim contained in claim 13; the second claim will be referred to as “the unnumbered claim”.
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, 7, 8, 10, 11, 16, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over CN 116188240 B (anonymous inventor) in view of Cowperthwaite et al. (US 20230306552 A1; hereinafter "Cowperthwaite").
Regarding claim 1, CN 116188240 B teaches: A resource control method comprising:
intercepting, by an interception module running in a kernel state, a system call request initiated by a process to a kernel driver in an operating system ([n0005] “In one possible implementation, the method further includes: intercepting a second IOCTL system call request initiated by the target container, wherein the second IOCTL system call request is used to request GPU memory”;
[n0046] states that “the judgment is performed in kernel mode”);
determining, by a configuration management module running in the kernel state, a target container running the process ([n0042] teaches that “…information is returned to the target container”, which requires identifying the target container which sent the request);
obtaining, by the configuration management module, a limited usage of GPU resources configured for the target container to use ([n0006] “In one possible implementation, the method further includes: when the amount of video memory requested does not exceed the amount of video memory mounted, calling the video memory resources of the physical GPU device for the target container based on the amount of video memory requested.”); and
configuring, by a resource management module running in the kernel state and based on the limited usage, an operation permission of the process to the target graphics memory buffer area ([n0056] “When the amount of video memory requested does not exceed the amount of video memory available, the GPU kernel module KMD is invoked based on the amount of video memory requested corresponding to the second IOCTL system call request to access the video memory resources of the physical GPU device for the target container.”).
CN 116188240 B does not explicitly teach: the system call request including a process ID of the process and address information of a target graphics memory buffer area that the process requests a GPU to allocate, or that the target container is determined based on the process ID.
Cowperthwaite teaches: the system call request including a process ID of the process and address information of a target graphics memory buffer area that the process requests a GPU to allocate (pg. 13 table 3 gives an example of information which may be contained in a system call, including a process ID and memory address; [0155] to [0160] explains further).
CN 116188240 B and Cowperthwaite are analogous to the claimed invention because they are in the same field of virtualized GPU resource management. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B with the teachings of Cowperthwaite to indicate a process ID and memory address in the system call. The motivation would have been to be able to provide a method for allocating specific memory to specific processes.
Regarding claim 7, the combination of CN 116188240 B in view of Cowperthwaite teaches: The method according to claim 1, further comprising:
constructing, by the interception module, a pseudo device file with a same name as a device file of the kernel driver in the operating system (CN 116188240 B: [n0036] “As shown in Figure 2, the virtual GPU device obtained through virtualization is renamed to the same name as the GPU resource of the physical GPU device, so that the target container in user space does not need to pay attention to the virtualization process of the physical GPU device, thus achieving seamless mounting of the virtual GPU device.”);
wherein intercepting the system call request includes obtaining the system call request from the pseudo device file (CN 116188240 B: fig. 2 and 3 show the IOCTL system call being sent to a virtual device before being passed to the physical device).
Regarding claim 8, the combination of CN 116188240 B in view of Cowperthwaite teaches: The method according to claim 1, further comprising, after determining the process ID and the address information:
forwarding, by the interception module, the system call request to the kernel driver (CN 116188240 B, request for memory is forwarded from the virtual GPU to the physical device at the kernel level: [n0055] “In one possible implementation, the method further includes: when the amount of video memory requested does not exceed the amount of video memory available, calling video memory resources of the physical GPU device for the target container based on the amount of video memory requested.”).
Claim(s) 2 and 12 is/are rejected under 35 U.S.C. 103 as being unpatentable over CN 116188240 B in view of Cowperthwaite (US 20230306552 A1) as applied to claims 1 and 11 above, and further in view of Zeng et al. (CN 111638953 A; hereinafter "Zeng").
Regarding claim 2, the combination of CN 116188240 B in view of Cowperthwaite teaches: The method according to claim 1, but does not explicitly teach: further comprising:
obtaining, by the resource management module, an actual usage of GPU resources currently used by the target container;
wherein configuring the operation permission includes:
configuring the operation permission based on the actual usage and the limited usage, the operation permission including allowing write operations or not allowing write operations.
Zeng teaches: obtaining, by the resource management module, an actual usage of GPU resources currently used by the target container ([0062] “During the implementation of application service requests, the actual GPU memory usage obtained from real-time monitoring of the application service process is used to determine whether it exceeds the GPU memory request amount for the application service request.”;
[0068] “The application service instance is essentially a process running in a container-based customized environment.”);
wherein configuring the operation permission includes:
configuring the operation permission based on the actual usage and the limited usage, the operation permission including allowing write operations or not allowing write operations ([0062] “If it does, a soft isolation mechanism is used to isolate the application service request, thereby protecting the normal operation of other application service requests on the same GPU.”).
Zeng is analogous to the claimed invention because it is in the same field of virtualized GPU resource management. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B in view of Cowperthwaite with the teachings of Zeng to allocate GPU resources based on the actual resource usage. The motivation would have been to improve efficiency by taking into account current demands in addition to theoretical needs.
Claim(s) 3, 5, 6, 13 (first claim - see 112 rejection section), 14, and 15 is/are rejected under 35 U.S.C. 103 as being unpatentable over CN 116188240 B in view of Cowperthwaite (US 20230306552 A1) and further in view of Zeng (CN 111638953 A) as applied to claims 2 and 12 above, and further in view of Chen et al. (CN 114168344 A; hereinafter "Chen").
Regarding claim 3, the combination of CN 116188240 B in view of Cowperthwaite and further in view of Zeng teaches: The method according to claim 2, but does not explicitly teach: further comprising:
in response to the operation permission being not allowing write operations, controlling the process to enter a dormant state responding to the resource management module detecting that the process initiates a write request to the target graphics memory buffer, the write request being used to request to write a GPU resource usage request.
Chen teaches: in response to the operation permission being not allowing write operations, controlling the process to enter a dormant state responding to the resource management module detecting that the process initiates a write request to the target graphics memory buffer, the write request being used to request to write a GPU resource usage request ([n0097] to [n0099] “The process of GPU soft isolation limiting GPU resource allocation includes: Determine if the current amount of free GPU resources is empty; If so, then postpone the execution of this GPU resource allocation;”).
Chen is analogous to the claimed invention because it is in the same field of virtualized GPU resource management. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B in view of Cowperthwaite and further in view of Zeng with the teachings of Chen to pause a process if it exceeds the predetermined GPU memory allocation. The motivation would have been to improve efficiency and ensure proper functioning of the system.
Regarding claim 5, the combination of CN 116188240 B in view of Cowperthwaite and further in view of Zeng teaches: The method according to claim 2, but does not explicitly teach: wherein the actual usage is a first actual usage;
the method further comprising:
in response to the operation permission being not allowing write operations, obtaining, by the resource management module, a second actual usage of the GPU resources currently used by the target container; and
in response to the second actual usage being less than the limited usage, configuring, by the resource management module, the operation permission to allowing write operations.
Chen teaches: wherein the actual usage is a first actual usage;
the method further comprising:
in response to the operation permission being not allowing write operations, obtaining, by the resource management module, a second actual usage of the GPU resources currently used by the target container; and
in response to the second actual usage being less than the limited usage, configuring, by the resource management module, the operation permission to allowing write operations (Chen [n0101] “Specifically, in the case of soft isolation, if the execution audit result returned by the eBPF process is that the audit fails, it is further determined whether the current amount of idle GPU resources is empty. If the amount of idle GPU resources is empty, the GPU resource allocation will be postponed.”;
where “postponed” may suggest to one of ordinary skill in the art that the method may be attempted again at a later time).
Chen is analogous to the claimed invention because it is in the same field of virtualized GPU resource management. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B in view of Cowperthwaite and further in view of Zeng with the teachings of Chen to resume a dormant process if its GPU usage decreases below the predetermined limit. The motivation would have been to allow a rendering task to be finished if possible rather than abandoning it.
Regarding claim 6, the combination of CN 116188240 B in view of Cowperthwaite and further in view of Zeng and Chen teaches: The method according to claim 5, further comprising:
in response to the second actual usage being less than the limited usage, waking up, by the resource management module, the process in a dormant state (Chen [n0101] “Specifically, in the case of soft isolation, if the execution audit result returned by the eBPF process is that the audit fails, it is further determined whether the current amount of idle GPU resources is empty. If the amount of idle GPU resources is empty, the GPU resource allocation will be postponed.”;
where “postponed” suggests resuming at a later time).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B in view of Cowperthwaite and further in view of Zeng with the teachings of Chen to resume a dormant process if its GPU usage decreases below the predetermined limit. The motivation would have been to allow a rendering task to be finished if possible rather than abandoning it.
Claim(s) s 9 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over CN 116188240 B in view of Cowperthwaite (US 20230306552 A1) as applied to claims 1 and 11 above, and further in view of Kim et al. (US 20200210242 A1; hereinafter "Kim") and An et al. (CN 114625482 A; hereinafter "An").
Regarding claim 9, the combination of CN 116188240 B in view of Cowperthwaite teaches: The method according to claim 1, but does not explicitly teach: further comprising:
in response to detecting that the target container is started, creating, by the configuration management module, a configuration file in a virtual file system, and mapping, by the configuration management module, a file path of the configuration file to a file path corresponding to a file for storing resource configuration information in the target container;
obtaining, by the configuration management module, the limited usage of the GPU resources that are able to be used by the target container configured in the configuration file; and
storing, by the configuration management module, a corresponding relationship between the limited usage and the target container.
Kim teaches: in response to detecting that the target container is started, creating, by the configuration management module, a configuration file in a virtual file system, and mapping, by the configuration management module, a file path of the configuration file to a file path corresponding to a file for storing resource configuration information in the target container ([0033] “Referring to FIG. 2, if a container is created (step S201), the node controller 121 transmits a configuration file including GPU resource constraint information and a system call/API profile to the container (step S203).”);
obtaining, by the configuration management module, the limited usage of the GPU resources that are able to be used by the target container configured in the configuration file ([0033] “…including GPU resource constraint information” (from previous citation)); and
storing, by the configuration management module, a corresponding relationship between the limited usage and the target container ([0033] “The library controller and the system call controller in the container may receive and store the configuration file including the resource constraint information.”).
Kim is analogous to the claimed invention because it is in the same field of virtualized GPU resource management. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B in view of Cowperthwaite with the teachings of Kim to store GPU resource allocation information associated with a target container in a configuration file. The motivation would have been to have an organized manner of keeping track of this information.
The combination of CN 116188240 B in view of Cowperthwaite and further in view of Kim does not explicitly teach creating a configuration file in a virtual file system in the kernel state.
An teaches mapping resource scheduling using configuration files at the kernel level ([n0055] “For example, after generating the mapping relationship, the resource scheduling system in the CPU can provide the mapping relationship to the vGPU runtime, which then virtualizes the GPU that the container can call based on the mapping relationship. The resource scheduling system can provide this mapping relationship to the vGPU runtime through configuration files, environment variables, command-line parameters, etc.”;
[n0062] “The first API is the API provided by the GPU manufacturer, which can be either a GPU user-space API or a GPU kernel driver API. The second API is an API provided by the vGPU runtime that has the exact same name, appearance, and other characteristics as the API provided by the GPU manufacturer. This second API is used to call the vGPU instance within each container.”)
An is analogous to the claimed invention because it is in the same field of virtualized GPU resource management. It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to have modified the invention of CN 116188240 B in view of Cowperthwaite and further in view of Kim with the teachings of An to create a configuration file in the kernel state. The motivation would have been to improve efficiency by performing operations in the kernel space.
Regarding claim 10, it is rejected with the same references, rationale, and motivation to combine as claim 1 because its limitations substantially correspond to the limitations of claim 1, as well as the additional limitation of: A resource control device (CN 116188240 B: [n0003] “This disclosure presents a technical solution for a GPU virtualization method and apparatus, and an electronic device for containers.”).
Regarding claims 11-13 and 14-18, they are rejected with the same references, rationale, and motivation to combine as claims 1-3 and 5-9 because their limitations substantially correspond to the limitations of claims 1-3 and 5-9 (note that “claim 13” refers to the first claim listed under the number 13 – see “Claim Rejections - 35 USC § 112” section), as well as the additional limitation of: An electronic apparatus comprising: a memory storing a computer program; and a processor configured to execute the computer program (CN 116188240 B: [n0012] “According to one aspect of this disclosure, an electronic device is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to invoke the instructions stored in the memory to perform the method described above.”).
Allowable Subject Matter
Claim 4 objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
The unnumbered claim following claim 13 would be allowable if rewritten to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action and to include all of the limitations of the base claim and any intervening claims.
The following is a statement of reasons for the indication of allowable subject matter:
The prior art does not teach or suggest the limitations of claim 4: “wherein controlling the process to enter the dormant state responding to the resource management module detecting that the process initiates the write request includes:
responding to a signal management module in the target container detects an operation-unauthorized instruction sent by the operating system kernel, writing a system call instruction into a target device file, the operation-unauthorized instruction indicating that the process initiates the write request to the target graphics memory buffer area while the operation permission is not allowing write operations, and the target device file being used for information exchange between a user state and kernel state; and
controlling, by the resource management module, the process to enter the dormant state in response to detecting that the system call instruction is written into the target device file.” – in particular, the limitations regarding the specifics of responding to the detection of an operation-unauthorized instruction sent by the operating system kernel.
Likewise, the prior art does not teach or suggest the limitations of the unnumbered claim following claim 13, which substantially correspond to the limitations of claim 4.
References Cited
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Cerny (US 20210096917 A1) teaches suspending a process if it exceeds its allotted GPU time slice.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BENJAMIN STATZ whose telephone number is (571)272-6654. The examiner can normally be reached Mon-Fri 8am-5pm.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tammy Goddard can be reached at (571)272-7773. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
BENJAMIN STATZ
Examiner
Art Unit 4167
/BENJAMIN TOM STATZ/ Examiner, Art Unit 2611
/TAMMY GODDARD/ Supervisory Patent Examiner, Art Unit 2611
Prosecution Insights