Abstract: Works in the literature fail to leverage embedding access patterns and memory units' access/storage capabilities, which when combined can yield high-speed heterogeneous systems by dynamically re-organizing embedding tables partitions across hardware during inference. A method and system for optimal deployment of embeddings tables across heterogeneous memory architecture for high-speed recommendations inference is disclosed, which dynamically partitions and organizes embedding tables across fast memory architectures to reduce access time. Partitions are chosen to take advantage of the past access patterns of those tables to ensure that frequently accessed data is available in the fast memory most of the time. Partition and replication is used to co-optimize memory access time and resources.

Abstract: The disclosure generally relates to an FPGA-based online 3D bin packing. Online 3D bin packing is the process of packing boxes into larger bins-Long Distance Containers (LDCs) such that the space inside each LDC is used to the maximum extent. The use of deep reinforcement learning (Deep RL) for this process is effective and popular. However, since the existing processor-based implementations are limited by Von-Neumann architecture and take a long time to evaluate each alignment for a box, only a few potential alignments are considered, resulting in sub-optimal packing efficiency. This disclosure describes an architecture for bin packing which leverages pipelining and parallel processing on FPGA for faster and exhaustive evaluation of all alignments for each box resulting in increased efficiency. In addition, a suitable generic purpose processor is employed to train the neural network within the algorithm to make the disclosed techniques computationally light, faster and efficient.

Abstract: Techniques are disclosed for reallocating host resources in a virtualized computing environment when certain criteria have been met. In some embodiments, a system identifies a host disabling event. In view of the disabling event, the system identifies a resource for reallocation from a first host to a second host. Based on the identification, the computer system disassociates the identified resource's virtual identifier from the first host device and associates the virtual identifier with the second host device. Thus, the techniques disclosed significantly reduce a system's planned and unplanned downtime.

Abstract: Methods and devices for providing reserved failover capacity across a plurality of data centers are described herein. An exemplary method includes determining whether a management process is executing at a first data center corresponding to a first physical location. In accordance with a determination that the management process is not executing at the first data center corresponding to the first physical location a host is initiated at a second data center corresponding to a second physical location and the management process is executed on the initiated host at the second data center corresponding to the second physical location.

Abstract: Processes for managing computing processes within a plurality of data centers configured to provide a cloud computing environment are described. An exemplary process includes executing a process on a first host of a plurality of hosts. When the process is executing on the first host, a first network identifier associated with the plurality of hosts is not a network identifier of a pool of network identifiers associated with the cloud computing environment and first and second route tables respectively corresponding to first and second data centers of the plurality of data centers associate the first network identifier with the first host. The exemplary process further includes detecting an event associated with the process. In response to detecting the event associated with the process, the first and second route tables are respectively updated to associate the first network identifier with a second host of the plurality of hosts.

Abstract: Processes for managing computing processes within a plurality of data centers configured to provide a cloud computing environment are described. An exemplary process includes executing a process on a first host of a plurality of hosts. When the process is executing on the first host, a first network identifier associated with the plurality of hosts is not a network identifier of a pool of network identifiers associated with the cloud computing environment and first and second route tables respectively corresponding to first and second data centers of the plurality of data centers associate the first network identifier with the first host. The exemplary process further includes detecting an event associated with the process. In response to detecting the event associated with the process, the first and second route tables are respectively updated to associate the first network identifier with a second host of the plurality of hosts.

Abstract: A system is provided for at least one of testing and validating at least one Open Radio Access Network (O-RAN) remote radio head (RRH) present for installation at an installation site having no network connectivity to the RRH available, which system includes a network emulator configured to at least one of test and validate at least one of a device parameter and a functional parameter of the at least one RRH when selectively coupled to the at least one RRH. The network emulator is configured to be selectively coupled to a user equipment (UE) configured for displaying the at least one of the device parameter and the functional parameter of at least one RRH, and wherein the network emulator is one of 1) directly coupled to the at least one RRH, or 2) connected to the at least one RRH via a cell site router (CSR).

Abstract: The present disclosure describes a technique for honoring virtual machine placement constraints established on a first host implemented on a virtualized computing environment by receiving a request to migrate one or more virtual machines from the first host to a second host and without violating the virtual machine placement constraints, identifying an architecture of the first host, provisioning a second host with an architecture compatible with that of the first host, adding the second host to the cluster of hosts, and migrating the one or more virtual machines from the first host to the second host.

Abstract: Processes for managing computing processes within a plurality of data centers configured to provide a cloud computing environment are described. An exemplary process includes executing a process on a first host of a plurality of hosts. When the process is executing on the first host, a first network identifier associated with the plurality of hosts is not a network identifier of a pool of network identifiers associated with the cloud computing environment and first and second route tables respectively corresponding to first and second data centers of the plurality of data centers associate the first network identifier with the first host. The exemplary process further includes detecting an event associated with the process. In response to detecting the event associated with the process, the first and second route tables are respectively updated to associate the first network identifier with a second host of the plurality of hosts.

Abstract: A number of hosts in a logical cluster is adjusted up or down in an elastic manner by tracking membership of hosts in the cluster using a first data structure and tracking membership of hosts in a spare pool using a second data structure, and upon determining that a triggering condition for adding another host is met and that all hosts in the cluster are being used, selecting a host from the spare pool, and programmatically adding an identifier of the selected host to the first data structure and programmatically deleting the identifier of the selected host from the second data structure.

Abstract: Processes for managing computing processes within a plurality of data centers configured to provide a cloud computing environment are described. An exemplary process includes executing a process on a first host of a plurality of hosts. When the process is executing on the first host, a first network identifier associated with the plurality of hosts is not a network identifier of a pool of network identifiers associated with the cloud computing environment and first and second route tables respectively corresponding to first and second data centers of the plurality of data centers associate the first network identifier with the first host. The exemplary process further includes detecting an event associated with the process. In response to detecting the event associated with the process, the first and second route tables are respectively updated to associate the first network identifier with a second host of the plurality of hosts.

Abstract: Methods and devices for providing reserved failover capacity across a plurality of data centers are described herein. An exemplary method includes determining whether a management process is executing at a first data center corresponding to a first physical location. In accordance with a determination that the management process is not executing at the first data center corresponding to the first physical location a host is initiated at a second data center corresponding to a second physical location and the management process is executed on the initiated host at the second data center corresponding to the second physical location.

Abstract: Techniques are disclosed for reallocating host resources in a virtualized computing environment when certain criteria have been met. In some embodiments, a system identifies a host disabling event. In view of the disabling event, the system identifies a resource for reallocation from a first host to a second host. Based on the identification, the computer system disassociates the identified resource's virtual identifier from the first host device and associates the virtual identifier with the second host device. Thus, the techniques disclosed significantly reduce a system's planned and unplanned downtime.

Abstract: Disclosed are aspects of proactive high availability that proactively identify and predict hardware failure scenarios and migrate virtual resources to healthy hardware resources. In some aspects, a mapping that maps virtual resources to hardware resources. A plurality of hardware events are identified in association with a hardware resource. A hardware failure scenario is predicted based on a health score of a first hardware resource. A health score is determined based on the hardware events, and a fault model that indicates a level of severity of the hardware events. A particular virtual resource is migrated from the hardware resource to another hardware that has a greater health score.

Abstract: Methods and devices for providing reserved failover capacity across a plurality of data centers are described herein. An exemplary method includes determining whether a management process is executing at a first data center corresponding to a first physical location. In accordance with a determination that the management process is not executing at the first data center corresponding to the first physical location a host is initiated at a second data center corresponding to a second physical location and the management process is executed on the initiated host at the second data center corresponding to the second physical location.

Abstract: Techniques are disclosed for reallocating host resources in a virtualized computing environment when certain criteria have been met. In some embodiments, a system identifies a host disabling event. In view of the disabling event, the system identifies a resource for reallocation from a first host to a second host. Based on the identification, the computer system disassociates the identified resource's virtual identifier from the first host device and associates the virtual identifier with the second host device. Thus, the techniques disclosed significantly reduce a system's planned and unplanned downtime.

Abstract: A system for monitoring a virtual machine executed on a host. The system includes a processor that receives an indication that a failure caused a storage device to be inaccessible to the virtual machine, the inaccessible storage device impacting an ability of the virtual machine to provide service, and applies a remedy to restore access to the storage device based on a type of the failure.

Abstract: The present disclosure is related to systems and methods for protecting virtual computing instances. An example system can include a first virtual computing instance (VCI) deployed on a hypervisor and provisioned with a pool of physical computing resources. The hypervisor and the first VCI can operate according to a first configuration profile. The system can include a fault domain manager (FDM) running on a second VCI that is deployed on the hypervisor and provisioned by the pool of physical computing resources. The FDM can be configured to provide high availability support for the first VCI, and the FDM can operate according to a second configuration profile. The system can further include a hypervisor manager running on the second VCI. The hypervisor manager can be configured to facilitate interaction between the FDM and the hypervisor by translating between the first configuration profile and the second configuration profile.

Abstract: Techniques are disclosed for reallocating host resources in a virtualized computing environment when certain criteria have been met. In some embodiments, a system identifies a host disabling event. In view of the disabling event, the system identifies a resource for reallocation from a first host to a second host. Based on the identification, the computer system disassociates the identified resource's virtual identifier from the first host device and associates the virtual identifier with the second host device. Thus, the techniques disclosed significantly reduce a system's planned and unplanned downtime.

Abstract: Methods and devices for providing reserved failover capacity across a plurality of data centers are described herein. An exemplary method includes determining whether a management process is executing at a first data center corresponding to a first physical location. In accordance with a determination that the management process is not executing at the first data center corresponding to the first physical location a host is initiated at a second data center corresponding to a second physical location and the management process is executed on the initiated host at the second data center corresponding to the second physical location.