公开(公告)号：US20230297425A1

公开(公告)日：2023-09-21

申请号：US18201010

申请日：2023-05-23

申请人： Imagination Technologies Limited

发明人： Luca Iuliano ,  Simon Nield ,  Yoong-Chert Foo ,  Ollie Mower ,  Jonathan Redshaw

IPC分类号： G06F9/50 ,  G06F12/084 ,  G06F12/02 ,  G06F12/0842 ,  G06F9/48 ,  G06F9/54

CPC分类号： G06F9/5016 ,  G06F12/084 ,  G06F12/0223 ,  G06F12/0842 ,  G06F9/4881 ,  G06F9/5022 ,  G06F9/505 ,  G06F9/544 ,  G06F12/02 ,  G06F12/00

摘要： A resource allocator receives a memory resource request for first memory resources in respect of a first-received task of a workgroup having a plurality of tasks. In response to receiving the memory resource request, the resource allocator allocates to the entire workgroup a block of memory portions of a shared memory that is sufficient in size for each task of the workgroup to receive memory resources in the block equivalent to the first memory resources.

公开(公告)号：US20230289208A1

公开(公告)日：2023-09-14

申请号：US18445068

申请日：2023-03-27

申请人： Google LLC

发明人： Dean Hildebrand ,  Grigor Avagyan

IPC分类号： G06F9/455 ,  G06F9/50

CPC分类号： G06F9/45558 ,  G06F9/5022 ,  G06F9/45554 ,  G06F2009/45587 ,  G06F2209/5011

摘要： A secondary pool of VMs is used to run secondary services or jobs, which may be evicted upon failure of a corresponding primary VM. Upon detection of a failure of a primary resource, the secondary services or jobs are evicted from secondary pool resources, and the secondary pool resources can be automatically allocated to the jobs of the failed primary resource. In this regard, a secondary job may be thought of as a preemptible job and comprises services or jobs that are lower priority than the service or job on the primary resource. By using computing resources in the secondary pool to run secondary or preemptible jobs, this technology makes use of what would be otherwise idle resources. This beneficially avoids having to allocate additional and separate computing resources for secondary jobs, leads to more efficient use of network resources, and reduces costs.

公开(公告)号：US20230244528A1

公开(公告)日：2023-08-03

申请号：US17591179

申请日：2022-02-02

申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

发明人： Wen Tao Zhang ,  Yiwen Huang ,  Yang Yang ,  Hong Jun Tu ,  Xiao Kai Dong ,  YE CUI ,  Wei Wu ,  Yu Chi Zhang ,  Xin Huang

IPC分类号： G06F9/50 ,  G06F8/65

CPC分类号： G06F9/5016 ,  G06F9/5022 ,  G06F9/5044 ,  G06F8/65

摘要： The invention discloses a method, system, and computer program product for service update management. The method, during a first update procedure for a first service in a computing system, in response to a detection that the first update procedure fails, a state of a second service on which the first service depends in the computing system is checked. In response to the second service being in a first state, the first update procedure is suspended. Resources allocated to the first update procedure by the computing system are released. The released resources are allocated to the second update procedure of the second service. The second update procedure is implemented based on the allocated released resources. With these embodiments, failures during the update procedures for the services may be automatically detected and removed based on a dependency relationship between the services. Therefore, the performance of the update procedures may be increased.

公开(公告)号：US11704200B2

公开(公告)日：2023-07-18

申请号：US17823877

申请日：2022-08-31

申请人： Snowflake Inc.

发明人： Ata E. Husain Bohra ,  Daniel Geoffrey Karp

IPC分类号： G06F11/14 ,  G06F16/245 ,  G06F16/25 ,  G06F9/50

CPC分类号： G06F11/1435 ,  G06F9/505 ,  G06F9/5022 ,  G06F9/5038 ,  G06F16/245 ,  G06F16/256

摘要： The subject technology retrieves information related to a set of instances of compute service managers, each instance of a particular compute service manager being associated with a set of virtual warehouses. The subject technology filters the information to determine a set of candidates from the set of instances of compute service managers. The subject technology sorts the set of candidates based at least in part on each workload of each of the set of candidates. The subject technology selects a candidate compute service manager from the set of instances of compute service managers to issue a query restart by randomly selecting an execution node, the execution node being included in a particular virtual warehouse associated with the candidate compute service manager, the selecting facilitating improving utilization of cluster resources and improving query execution on the selected candidate compute service manager.

公开(公告)号：US20230222004A1

公开(公告)日：2023-07-13

申请号：US17572122

申请日：2022-01-10

申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

发明人： Rui Wang ,  Jing Bo Jiang ,  Yi Ming Wang ,  Yan Liu

IPC分类号： G06F9/50 ,  G06N7/00

CPC分类号： G06F9/5077 ,  G06F9/5083 ,  G06F9/5038 ,  G06F9/5022 ,  G06N7/005 ,  G06F2209/508

摘要： Controlling data locality in a Kubernetes computing environment by establishing a Kubernetes computing environment including a controller and at least one executor pod for running an application, and receiving a request for a task to be run in the Kubernetes computing environment. The controller dispatches a sidecar to collect resource data from the at least one executor pod for an input to a directed acyclic graph (DAG) feature analyzer. The directed acyclic graph (DAG) feature analyzer identifies from the at least one executor pod a best dynamic resource that are available to execute. The at least one executor pod meeting the best dynamic resource that is available executes the task to be run in the Kubernetes computing.

公开(公告)号：US20230185613A1

公开(公告)日：2023-06-15

申请号：US18168049

申请日：2023-02-13

申请人： Samsung Electronics Co., Ltd.

发明人： Jaehong Park

IPC分类号： G06F9/50 ,  G06F15/02 ,  G06F11/30 ,  G06F11/34

CPC分类号： G06F9/5016 ,  G06F9/5022 ,  G06F11/3037 ,  G06F11/3075 ,  G06F11/3089 ,  G06F11/3409 ,  G06F15/0208 ,  G06F9/4862 ,  G06F2201/81 ,  G06F2209/508 ,  G06F2209/5022

摘要： An electronic device including an application processor and a communication processor. The communication processor, for managaging a resource memory, configured to monitor an occupancy rate of the resource memory for detecting a memory leakage, release a network connection when the electronic device is in an idle state, in response to the detecting of the memory leakage, initialize the resource memory after the network connection is released, and reconnect the network connection, wherein the idle state indicates a state when the electronic device is not occupied or used by a user, and wherein the initialization is performed at a non-flag area from among the resource memory.

公开(公告)号：US20230185612A1

公开(公告)日：2023-06-15

申请号：US17546643

申请日：2021-12-09

申请人： NVIDIA Corporation

发明人： Vivek Belve Kini ,  Steven Arthur Gurfinkel ,  Fnu Vishnuswaroop Ramesh ,  Houston Thompson Hoffman ,  Michael Christopher Delorme ,  Alicia Xiao Hu ,  Stephen Anthony Bernard Jones ,  Vladislav Zhurba ,  William Young Fiser ,  Reza Mokhtari

IPC分类号： G06F9/50 ,  G06F9/38

CPC分类号： G06F9/5016 ,  G06F9/5022 ,  G06F9/5044 ,  G06F9/3885 ,  G06F2209/5011

摘要： Apparatuses, systems, and techniques to allocate processor memory. In at least one embodiment, an application programming interface is used to execute instructions to asynchronously allocate memory locations to processors.

公开(公告)号：US11675625B2

公开(公告)日：2023-06-13

申请号：US17116988

申请日：2020-12-09

申请人： Dell Products L.P.

发明人： William Price Dawkins ,  Robert Wayne Hormuth ,  Walter A. O'Brien, III ,  Yossef Saad ,  Mukund P. Khatri ,  Gaurav Chawla ,  Mark Steven Sanders ,  Elie Antoun Jreij ,  Jimmy Doyle Pike

IPC分类号： G06F9/46 ,  G06F9/50 ,  G06F11/30

CPC分类号： G06F9/505 ,  G06F9/5022 ,  G06F9/5044 ,  G06F9/5077 ,  G06F11/3051 ,  G06F2209/508

摘要： A method for managing composed information handling systems includes obtaining, by a system control processor, a composition request for a composed information handling system of the composed information handling systems, wherein the composition request comprises an operational instances requirement, and initially allocating only a minimum number of instances of a resource based on the operational instances requirement to the composed information handling system, wherein at least one of the minimum number of instances of the resource is provided by a first information handling system, wherein the first information handling system is operatively connected to the system control processor manager.

公开(公告)号：US20230153155A1

公开(公告)日：2023-05-18

申请号：US17525716

申请日：2021-11-12

申请人： Amazon Technologies, Inc.

发明人： Milan Krneta ,  Eric M. Kessler ,  Christian Bruun Madsen

IPC分类号： G06F9/50 ,  G06N10/00

CPC分类号： G06F9/5022 ,  G06F9/5044 ,  G06F9/5077 ,  G06N10/00

摘要： An algorithm execution management system of a provider network may receive a request from a user for executing an algorithm using different types of computing resources, including classical computing resources and quantum computing resources. The request may indicate a container that includes the algorithm code and dependencies such as libraries for executing the algorithm. The algorithm execution management system may first determine that the quantum computing resources are available to execute the algorithm, and then cause the classical computing resources to be provisioned. The algorithm execution management system may cause at least one portion of the algorithm to be executed at the classical computing resources using the container indicated by the user, and at least another portion of the algorithm to be executed at the quantum computing resources. The quantum task of the algorithm may be provided a priority during execution of the algorithm for using the quantum computing resources.

公开(公告)号：US11645145B2

公开(公告)日：2023-05-09

申请号：US16716247

申请日：2019-12-16

申请人： QUALCOMM Incorporated

发明人： Andrew Evan Gruber

IPC分类号： G06F11/07 ,  G06T1/60 ,  G06F9/30 ,  G06F9/38 ,  G06F9/50 ,  G06F9/54 ,  G06F12/0871 ,  G06F12/0882 ,  G09G5/00

CPC分类号： G06F11/0793 ,  G06F9/30047 ,  G06F9/3861 ,  G06F9/5016 ,  G06F9/5022 ,  G06F9/544 ,  G06F11/0772 ,  G06F12/0871 ,  G06F12/0882 ,  G06T1/60 ,  G09G5/001

摘要： The present disclosure relates to methods and apparatus for display processing. For example, disclosed techniques facilitate speculative page fault handling in a GPU. Aspects of the present disclosure can perform a graphics operation associated with using a set of constants within a flow control. Aspects of the present disclosure can also query a first memory to determine whether memory addresses associated with the set of constants are allocated at a constant buffer of the first memory. Further, aspects of the present disclosure can set a page fault indicator to a true value when the query indicates that at least one memory address associated with the set of constants is unallocated at the constant buffer, and set the page fault indicator to a false value otherwise.