Abstract: Embodiments of the present disclosure provide methods, devices, and computer program products for processing a task. A method of processing a task comprises: receiving, at a network device and from a set of computing devices, a set of processing results derived from processing the task by the set of computing devices; in response to receiving the set of processing results, executing a reduction operation on the set of processing results; and transmitting a result of the reduction operation to the set of computing devices. In this way, embodiments of the present disclosure can significantly reduce an amount of data exchanged among a plurality of devices processing a task in parallel, and thus reduce network latency caused by data exchange.

Abstract: Embodiments provide a method and machine for application migration. In an embodiment, a first descriptor associated with a first runtime environment of an application executed on a first dedicated processing unit is obtained from the first dedicated processing unit. In response to determining that the application is to be migrated from the first dedicated processing unit to the second dedicated processing unit during the execution, the second dedicated processing unit is caused to create a second descriptor associated with a second runtime environment for the application, executed on the second dedicated processing unit. Subsequently, the second descriptor is obtained from the second dedicated processing unit and a first mapping from the first descriptor to the second descriptor is established for re-directing, to the second dedicated processing unit, the request from the application to the first dedicated processing unit. A corresponding machine for implementing the method is further disclosed.

Abstract: Embodiments of the present disclosure relate to a method of tracing a computing system, a device for tracing a computing system, and a computer readable medium. According to some embodiments, tracing data is extracted from a request that requests a dedicated processing resource for a task, the request being initiated by an application executed on a client and the tracing data including a parameter for performing the task, an identifier of the application, and time elapsed from initiating the request. The tracing data is stored in a volatile memory to facilitate transmitting the tracing data to a database server. The request is caused to be processed by a computing server hosting the dedicated processing resource. In this way, the cloud computing system may be traced rather than tracing the stand-alone tasks only.

Abstract: Systems and methods are provided for implementing an intelligent data management system for data storage and data management in a cloud computing environment. For example, a system includes an application server, a distributed data storage system, and an intelligent data management system. The application server is configured to host a data processing application. The distributed data storage system is configured to store data generated by a network of devices associated with the data processing application. The intelligent data management system is configured to manage data storage operations for storing the data generated by the network of devices in the distributed data storage system. For example, the intelligent data management system is configured to determine one or more data types of the data generated by the network of devices and select one of a plurality of repositories within the distributed data storage system to store the data based on the determined data types.

Abstract: A method comprises: compiling the code segment with a compiler; and determining, based on an intermediate result of the compiling, a resource associated with a dedicated processing unit and for executing the code segment. As such, the resource required for executing a code segment may be determined quickly without actually executing the code segment and allocating or releasing the resource, which helps subsequent resource allocation and further brings about a better user experience.

Abstract: This invention provides for a compound of formula (I): or a pharmaceutically acceptable salt thereof, wherein R1-R4, L and X are defined herein. The compounds of formula (I) and pharmaceutically acceptable salts thereof are cationic lipids useful in the delivery of biologically active agents to cells and tissues.

Abstract: Embodiments of the present disclosure relate to a method, a device and a computer readable medium for managing a dedicated processing resource. According to the embodiments of the present disclosure, a server receives a request of a first application from a client, and based on an index of a resource subset as comprised in the request, determines a dedicated processing resource corresponding to the resource subset for processing the first application request. According to the embodiments of the present disclosure, the dedicated processing resource is divided into a plurality of resource subsets, so that the utilization efficiency of the dedicated processing resource is improved.

Abstract: Techniques are provided to decouple data pipeline tasks from an execution flow of a high-performance computing task (e.g., distributed deep model training) in a distributed computing system. For example, a method includes receiving a client request to provision resources for executing a computing job, provisioning accelerator resources of one or more accelerator server nodes in the distributed computing system to perform tasks associated with an execution flow of the computing job, and provisioning a logical nodes within the distributed computing system to compose a data flow pipeline which is configured to perform data flow operations associated with the computing job for providing data to the provisioned accelerator resources to perform the tasks associated with the execution flow of the computing job. The data flow operations include, e.g., data storage input/output operations, data pre-processing operations, and data staging operations, which are decoupled from the execution flow of the computing job.

Abstract: The present disclosure relate to a method and an electronic device for application migration.

Abstract: Embodiments of the present disclosure relate to a method, device and computer program product for deploying a machine learning model. The method comprises: receiving an intermediate representation indicating processing of a machine learning model, learning parameters of the machine learning model, and a computing resource requirement for executing the machine learning model, the intermediate representation, the learning parameters, and the computing resource requirement being determined based on an original code of the machine learning model, the intermediate representation being irrelevant to a programming language of the original code; determining, at least based on the computing resource requirement, a computing node and a parameter storage node for executing the machine learning model; storing the learning parameters in the parameter storage node; and sending the intermediate representation to the computing node for executing the machine learning model with the stored learning parameters.

Abstract: This invention provides for a compound of formula (I): or a pharmaceutically acceptable salt thereof, wherein R1-R4, L and X are defined herein. The compounds of formula (I) and pharmaceutically acceptable salts thereof are cationic lipids useful in the delivery of biologically active agents to cells and tissues.

Abstract: Embodiments of the present disclosure provide a method of managing dedicated processing resources, a server system and a computer program product. The method may include receiving a request for the dedicated processing resources from an application having an assigned priority. The method may also include determining a total amount of resources to be occupied by the application based on the request. The method may further include, in response to the total amount of resources approximating or exceeding a predetermined quota associated with the priority, assigning, from the dedicated processing resources, a first amount of dedicated processing resources to the application. Besides, the method may include, in response to the application completing an operation associated with the request using the assigned dedicated processing resources, causing the application to sleep for a time period.

Abstract: Techniques are provided for topology-aware provisioning of computing resources in a distributed heterogeneous environment.

Abstract: A method, apparatus and computer program product for managing memories of computing resources is disclosed. In the method, a computing task processed by a first computing resource in a group of computing resources is determined. In response to a second memory of a second computing resource other than the first computing resource in the group of computing resources being allocated to the computing task, a second access speed with which the first computing resource accesses the second memory is determined. A target computing resource is selected from the group of computing resources based on an access speed with which the first computing resource accesses a target memory of the target computing resource, where the access speed is higher than the second access speed. At least one part of data in the second memory is migrated to the target memory.

Abstract: Embodiments of the present disclosure provide a method, device and computer for managing a memory of a dedicated processing resource. The method determining, in a process of storing data for a computing task of a first dedicated processing resource of a set of dedicated processing resources to the first dedicated processing resource, a size of an available space of a memory of the first dedicated processing resource; in response to the size of the available space of the memory of the first dedicated processing resource being lower than a predetermined threshold value, determining a second dedicated processing resource of the set of dedicated processing resources, a size of an available space of a memory of the second dedicated processing resource is greater than the predetermined threshold value; and causing at least one portion of the data not stored on the memory of the first dedicated processing resource to be stored on the memory of the second dedicated processing resource.

Abstract: Embodiments of the present disclosure relate to a method, device and computer program product for storing data. The method comprises obtaining a first range of replica levels supported by a storage apparatus, wherein the replica level indicates the number of replicas of data. The method further comprises receiving a replica configuration requirement for an application, wherein the application supports a second range of replica levels. Moreover, the method further comprises determining a first replica level for the storage apparatus and a second replica level for the application based on the replica configuration requirement, the first range and the second range. By extracting the replica function supported by the storage apparatus, embodiments of the present disclosure can configure replica levels of the storage apparatus and the application from the global level for the user requirement for replicas of the data service.

Abstract: An apparatus in one embodiment comprises a host device that includes at least one processor and an associated memory. The host device is configured to implement a plurality of containers each configured to access one or more portions of the memory. The containers are implemented as respective kernel control groups utilizing operating system level virtualization of the processor of the host device. The host device is further configured to assign the containers to groups in accordance with one or more designated criteria, and to generate checkpoints for respective groups of the containers. In conjunction with generation of a given one of the checkpoints for a particular one of the groups of containers, the host device identifies one or more pages of the memory that are shared by at least first and second containers of the particular group of containers, and generates the given checkpoint without duplicating the one or more shared pages to persistent storage.

Abstract: At least one performance metric is monitored in an information processing system to detect a performance event substantially contemporaneous with the occurrence of the event, wherein monitoring of the performance metric is performed in a continuous manner. At least a portion of the information processing system is profiled in response to detection of a performance event, wherein the profiling step obtains a performance snapshot of the portion of the information processing system being profiled. The performance snapshot is analyzed to determine a root cause of the performance event within the information processing system.

Abstract: An apparatus in one embodiment comprises a plurality of container host devices of at least one processing platform. The container host devices implement a plurality of containers for executing applications on behalf of one or more tenants of cloud infrastructure. The containers are separated into at least first and second distinct container groups based at least in part on results of comparisons of layer structures each characterizing container images of respective different ones of the containers. At least one grouping-based container management action is determined for at least one of the containers based at least in part on which of the container groups includes that container. The grouping-based container management action is applied to one or more of the containers of one of the first and second container groups but not to the containers of the other of the first and second container groups.

Abstract: Embodiments of the present disclosure relates to a method and device of data access. The method comprises determining whether target data stored in a non-volatile storage device is cached in a memory. The target data is organized in a first level of a multi-way tree in the storage device. The method further comprises, in response to determining that the target data is missing in the memory, moving the target data from the storage device into the memory. Besides, the method comprises, in response to the target data being accessed from the memory, adding a reference to the target data to a first list, the first list recording a sequence for accessing data in the first level.