Abstract: Particular embodiments described herein provide for an electronic device that can be configured to receive a remote direct memory access (RDMA) message from a first virtual machine located on a first network element, determine that the RDMA message is destined for a second virtual machine that is located on the first network element, and use a local direct memory access engine to process the RDMA message, where the local direct memory access engine is located on the first network element. In an example, the electronic device can be further configured to determine that the RDMA message is destined for a third virtual machine on a second network element, wherein the second network element is different than the first network element and use an other device acceleration driver to process the RDMA message instead of the local direct memory access engine.

Abstract: Examples described herein relate to accessing an initiator as a Non-Volatile Memory Express (NMVe) device. In some examples, the initiator is configured with an address space, configured in kernel or user space, for access by a virtualized execution environment. In some examples, the initiator to copy one or more storage access commands from the virtualized execution environment into a queue for access by a remote direct memory access (RDMA) compatible network interface. In some examples, the network interface to provide Non-Volatile Memory Express over Fabrics (NVMe-oF) compatible commands based on the one or more storage access commands to a target storage device. In some examples, the initiator is created as a mediated device in kernel space or user space of a host system. In some examples, configuration of a physical storage pool address of the target storage device for access by the virtualized execution environment occurs by receipt of the physical storage pool address in a configuration command.

Abstract: Examples described herein relate to executing a poller to poll for received communications over multiple transport layer protocols from a connection to identify a received communication from one of the multiple transport layer protocols and identify a second received communication from a different one of the multiple transport layer protocols. A change to the different one of the multiple transport layer protocols occurs in response to failure of the one of the multiple transport layer protocols or slow transport rate using the one of the multiple transport layer protocols. In some examples, the poller is executed in user space and transport layer protocol processing of the received communication and second received communication occur in kernel space.

Abstract: Examples described herein relate to at least one processor that can execute a polling group to poll for storage transactions associated with a first group of one or more particular queue identifiers, wherein the one or more particular queue identifiers are associated with one or more queues that can be accessed using the polling group and no other polling group. In some examples, the polling group is to execute on a processor that runs no other polling group. In some examples, the at least one processor is configured to: execute a second polling group on a second processor, wherein the second polling group is to poll for storage transactions for a second group of one or more particular queue identifiers that are different than the one or more particular queue identifiers of the first group, wherein the second group of one or more particular queue identifiers are associated with one or more queues that can be accessed using the second polling group and not the first polling group.

Abstract: Technologies for hybrid acceleration of code include a computing device (100) having a processor (120), a field-programmable gate array (FPGA) (130), and an application-specific integrated circuit (ASIC) (132). The computing device (100) offloads a service request, such as a cryptographic request or a packet processing request, to the FPGA (130). The FPGA (130) performs one or more algorithmic tasks of an algorithm to perform the service request. The FPGA (130) determines one or more primitive tasks associated with an algorithm task and encapsulates each primitive task in a buffer that is accessible by the ASIC (132). The ASIC (132) performs the primitive tasks in response to encapsulation in the buffer, and the FPGA (130) returns results of the algorithm. The primitive operations may include cryptographic primitives such as modular exponentiation, modular multiplicative inverse, and modular multiplication.

Abstract: Examples herein relate to polling for input/output transactions of a network interface or a storage device, or any peripheral device. Some examples monitor clock cycles spent checking for a presence of input/output (I/O) events and processing I/O events and monitor clock cycles spent checking for presence of I/O events without completing an I/O event. Central processing unit (CPU) core utilization can be based on clock cycles spent checking for a presence of I/O events and processing I/O events and clock cycles spent checking for presence of I/O events without completion of an I/O event. For example, if core utilization is below a threshold, frequency of the core can be reduced for performing polling of I/O events. If core utilization is at or above the threshold, frequency of the core can be increased used to performing polling of I/O events.

Abstract: Examples described herein relate to allocation of power to engines and a surplus power available for use by one or more engines. A power management controller reserves power for one or more processors of the offload engine and provides a surplus power level. Based on a request for a workload performance by the first processor, the power management controller provides power to a processor based on the reserved power for the processor. Allocation of the surplus power can be made based on a priority of a workload relative to other workloads.

Abstract: Embodiments of the present invention relate to a method and apparatus for providing computing resources to a user. In one embodiment of the present invention, there is provided a method for providing computing resources to a user, comprising: in response to a request from the user, looking up in a profile repository a profile corresponding to the user's domain information; requesting at least one computing resource associated with the profile; integrating the at least one computing resource into a virtual datacenter; and providing the virtual datacenter to the user; wherein the at least one computing resource is physically deployed in a distributed manner. In one embodiment of the present invention, there is provided an apparatus for providing computing resources to a user.

Abstract: The present disclosure relates to a method and apparatus for measuring performance of a storage system. The method comprises: causing one or more entities to execute a task set comprising multiple tasks, each of the multiple tasks being used for accessing the storage system; obtaining an indicator set of the storage system based on a result of the execution, the indicator set comprising one or more indicators for indicating performance of the storage system; and adjusting the task set based on the indicator set, for subsequent execution by the one or more entities. The method can be executed iteratively. By means of the present invention, workloads for the next round's execution can be intelligently improved according to execution results after each round's execution of workloads, so that performance of the storage system can be obtained more pertinently and efficiently so as to better utilize the storage system.

Abstract: The present invention relates to a method and apparatus for data management in multi-tenant distributive environment. According to the embodiments of the present invention, a tenant-specific virtual metadata server may be created and registered with a task manager in a job processing system. The virtual metadata server provides a tenant-specific logical file system namespace. Any access to tenant data will be performed via the virtual metadata server. In other words, such virtual metadata server acts as an interface between a tenant terminal and an entity file system. In this way, it may effectively realize isolation of tenant data and enhance data security. Moreover, by using a virtual metadata server rather than physically isolating user data, the embodiments of the present invention may guarantee the efficiency of data sharing while improving the data security.

Abstract: A data storage system includes: one or more storage nodes configured to attach to a computer network to provide network-attached storage for the computer network; a first application programming interface providing data storage functionality at a file level in the one or more network-attached storage nodes for a first set of application programs operating in the computer network; and a second application programming interface providing data storage functionality at a level other than the file level in the one or more network-attached storage nodes for a second set of application programs operating in the computer network, wherein the second application programming interface comprises one or more native representational state transfer-based application programming interfaces.

Abstract: Techniques for profiling input/output (I/O) characteristics in a computing system implemented with distributed infrastructure. In one example, a method comprises the following steps. Behavioral characteristics are extracted for a set of factors associated with a targeted system in a distributed infrastructure by causing execution of one or more input/output workloads in accordance with one or more entities and the targeted system. At least a portion of the extracted behavioral characteristics is utilized to determine whether one or more subsequent workloads satisfy one or more quality-of-service criteria when executed in accordance with at least one of the one or more entities and the targeted system.

Abstract: Techniques for constructing virtual storage networks for tenants with quality-of-service delivery. In one example, a method comprises the following steps. One or more virtual storage networks are constructed respectively for one or more tenants of a data storage system. Each of the one or more virtual storage networks is tenant-managed and is configured such that logical resources of the tenant-managed virtual storage network are isolated from physical resources used to implement the logical resources.

Abstract: A first portion of a bootstrapping process is performed for a data storage system. The first portion of the bootstrapping process comprises a hypervisor scanning for and initializing a plurality of physical storage devices of the data storage system, and storing boot information obtained during the scanning and initializing steps. A second portion of a bootstrapping process is performed for the data storage system. The second portion of the bootstrapping process comprises providing at least a portion of the stored boot information to at least one virtual machine implementing an application program that utilizes the plurality of physical storage devices, and the virtual machine initializing one or more virtual resources using the stored boot information provided by the hypervisor. The scanning for the plurality of physical storage devices of the data storage system may be performed in a parallel manner.

Abstract: Techniques for delivering and measuring storage quality-of-service to virtual machines in a distributed virtual infrastructure. In one example, a method comprises the following steps. A controller obtains quality-of-service data from at least a portion of components of a distributed virtual infrastructure, wherein the components of the distributed virtual infrastructure comprise one or more storage units, one or more processing units, and one or more switches operatively coupled to form the distributed virtual infrastructure. The controller manages at least one input/output request throughout a path defined by at least one of the one or more processing units, at least one of the one or more switches, and at least one of the one or more storage units, based on at least a portion of the collected quality-of-service data.

Abstract: A resource request is sent from a user to a resource controller. The resource controller is resident on a node of a distributed computing platform comprising a plurality of nodes configured to collectively execute at least one application program. A quota pool associated with the resource controller is searched for a matching resource quota for the user to inherit. The resource quota is assigned to the user when a matching resource quota for the user to inherit is found in the quota pool. The resource request is forwarded to a resource scheduler that globally manages resources for the distributed computing platform, when a matching resource quota for the user to inherit is not found in the quota pool.

Abstract: Techniques are disclosed for managing assets, such as virtual assets, in a computing system implemented with distributed virtual infrastructure. In one example, a method comprises the following steps. Operational information associated with a plurality of virtual assets in a data center is obtained in a trusted manner. The data center is implemented via a distributed virtual infrastructure. At least a portion of the operational information for at least a portion of the plurality of virtual assets in the data center is reported. The operational information reported is operational information pertaining to one or more virtual assets that the data center provides for a tenant of the data center. The obtaining and reporting steps are performed by at least one processing device operating as a virtual asset manager operatively coupled to the distributed virtual infrastructure.

Abstract: A data storage system includes at least one network virtualization-aware switch. The data storage system also includes at least one storage array operatively coupled to the network virtualization-aware switch, wherein the storage array is operatively coupled to at least one host computing device, and wherein the host computing device instantiates at least one virtual machine in at least one virtual network. Further, the data storage system includes at least one controller operatively coupled to the network virtualization-aware switch, wherein the controller is configured to link the storage array with the virtual machine of the virtual network through the network virtualization-aware switch.

Abstract: Example embodiments of the present invention relate to a system, apparatus and methods for preserving the integrity of a code to prevent it from being modified, maliciously or inadvertently, while it is in execution in the RAM of a computer platform. This method also may be referred to as code hardening. Code to be hardened in example embodiments of the present invention may be referred to as protected code. Example embodiments of the present invention are able to externally detect unauthorized stoppage of the hypervisor by employing (1) a launch-time metric of the protected code; (2) a run-time metric of the protected code; and (3) a liveliness indicator of the protected code.

Abstract: Techniques for measuring input/output (I/O) scheduling characteristics in a computing system implemented with distributed virtual infrastructure. In one example, a method comprises the following steps. A plurality of virtual machines is deployed in a computing system implemented in a distributed virtual infrastructure managed by at least one hypervisor component. At least one test workload is executed on the plurality of virtual machines in accordance with the at least one hypervisor component. Timing information is collected from the plurality of virtual machines during execution of the at least one test workload. Based on at least a portion of the collected timing information, one or more characteristics are identified of an input/output schedule employed by the at least one hypervisor component during execution of the at least one test workload on the plurality of virtual machines.