Scheduling and clock management for real-time system quality of service (QoS) is disclosed. In an implementations, a resource manager determines a target work rate based on respective job deadlines of a plurality of jobs on a processing platform. Determining the target work rate can include ordering the plurality of jobs based on the respective deadlines, determining an amount of work required to reach each of the respective deadlines, identifying one deadline among the respective deadlines as a most constraining deadline based on the amount of work required to reach that one deadline, and determining the target work rate based on the most constraining deadline. The resource manager adjusts a clock rate of the processing platform based on at least the target work rate.

The technology disclosed herein enables computing node allocation based on build process specifications. An example method comprises receiving operational resource requirements of a software build job in a waiting state; identifying, for each computing node of a plurality of computing nodes associated with one or more other software build jobs, a respective set of operational resources installed on the computing node; identifying, among the plurality of computing nodes, a computing node having a minimal, among the plurality of computing nodes, difference between a set of operational resources installed in the computing node and the operational resource requirements of the software job; and scheduling the software build job to execute on the computing node.

Systems and methods described herein are directed to upgrading one or more of add-on firmware and disk firmware for a server, which can involve connecting a port of the server to an isolated network, the isolated network dedicated to firmware upgrades for the server; caching onto cache memory of the server, an operating system received through the isolated network; booting the operating system on the server from the cache memory; conducting an Network File System (NFS) mount on the server to determine hardware information associated with the upgrading of the one or more of the add-on firmware and the disk firmware; and upgrading the one or more of the add-on firmware and the disk firmware based on the hardware information.

Techniques are described for distributing network device tasks across virtual machines executing in a computing cloud. A network device includes a network interface to send and receive messages, a routing unit comprising one or more processors configured to execute a version of a network operating system, and a virtual machine agent. The virtual machine agent is configured to identify a virtual machine executing at a computing cloud communicatively coupled to the network device, wherein the identified virtual machine executes an instance of the version of the network operating system, to send, using the at least one network interface and to the virtual machine, a request to perform a task, and to receive, using the at least one network interface and from the virtual machine, a task response that includes a result of performing the task. The routing unit is configured to update the network device based on the result.

A system and method for exposing a cloud application programming interface (API) based on supported hardware of a cloud service system. The method includes executing a cloud service application to provide one or more services. The method includes maintaining, in a data storage, a plurality of linkages between a plurality of service identifiers corresponding to the one or more services of the cloud service application and a plurality of hardware identifiers. The method includes determining, based on the plurality of linkages and/or a hardware profile of the cloud service system, whether the cloud service system has a capability to provide access to the one or more services of the cloud service application. The method includes providing exposure to application programming interfaces (APIs) of at least one of the one or more services of the cloud service application.

Embodiments provide a thread classification method that represents stack traces in a compact form using classification signatures. Some embodiments can receive a stack trace that includes a sequence of stack frames. Some embodiments may generate, based on the sequence of stack frames, a trace signature that represents the set. Some embodiments may receive one or more subsequent stack traces. For each of the one or more subsequent stack traces, some embodiments may determine whether a subsequent trace signature has been generated to represent the sequence of stack frames included within the subsequent stack trace. If not, some embodiments may generate, based on the trace signature and other subsequent trace signatures that were generated based on the trace signature, the subsequent trace signature to represent the subsequent sequence of stack frames.

A cloud service system and an operation method thereof are provided. The cloud service system includes a first computing resource pool, a second computing resource pool, and a task dispatch server. Each computing platform in the first computing resource pool does not have a co-processor. Each computing platform in the second computing resource pool has at least one co-processor. The task dispatch server is configured to receive a plurality of tasks. The task dispatch server checks a task attribute of a task to be dispatched currently among the tacks. The task dispatch server chooses to dispatch the task to be dispatched currently to the first computing resource pool or to the second computing resource pool for execution according to the task attribute.

A gateway for use in a computing system to interface a host with the subsystem for acting as a work accelerator to the host, the gateway having: an accelerator interface for connection to the subsystem to enable transfer of batches of data between the subsystem and the gateway; a data connection interface for connection to external storage for exchanging data between the gateway and storage; a gateway interface for connection to at least one second gateway; a memory interface connected to a local memory associated with the gateway; and a streaming engine for controlling the streaming of batches of data into and out of the gateway in response to pre-compiled data exchange synchronisation points attained by the subsystem, wherein the streaming of batches of data are selectively via at least one of the accelerator interface, data connection interface, gateway interface and memory interface.

Compressed verbatim copy can enable more efficient copying of compressed data. In one example, a compressed verbatim copy method involves receiving a command to copy compressed data from a source address of the memory device to a destination address. In response to the receipt of the command, the method involves copying the compressed data in a compressed format from the source address to the destination address without first decompressing the data. A second source address and a second destination address of metadata for the compressed data is determined, and the metadata is copied from the second source address to the second destination address.

Peer-to-peer arrangements between graphics processing units (GPUs) are provided herein. A method includes establishing synthetic devices representing GPUs in an address domain associated with a host processor, where the GPUs have a different address domain than the host processor. The method also includes forming a peer arrangement between the GPUs such that data transfers between the GPUs in the different address domain can be initiated by the host processor interfacing with the synthetic devices.