Dell Products Inc.

Qualcomm Incorporated

Walmart Apollo, LLC

Continental Automotive Gmbh

Ankit Kumar Singh

Deepaganesh Paulraj

Sumanth Vidyadhara

Vaideeswaran Ganesan

Vinod Parackal Saby

Rohit Kumar Arehalli

Shekar Babu Suryanarayana

Vinod P S

Shrikant U Hallur

Kalyan Kumar Bhiravabhatla

Andrew Evan Gruber

Prasoon Kumar Sinha

Aniruddha Suresh Herekar

Shekar Babu S

Vishwanath Shashikant Nikam

Pavan Kumar Akkaraju

Krishnaiah Gummidipudi

Rohitkumar Arehalli

James Peter Giannoules

Shinose Abdul Rahiman

Jonnala Gadda Nagendra Kumar

Srihari Babu Alla

Manjunath Vishwanath

Ajay Shenoy

Naveen Karthick Chandrasekaran

Yasaswi Saisriram Bhimaraju

Guru Prasad Yadav

Neeraj Kumar Pant

Sreenivasula Reddy G

Abhishek Lal

Suvam Chatterjee

Sharad Raj

Renju Boben

Sampathkumar Periasamy

Robin Mathew

S Shekar Babu

Shekar Babu Suryananarayana

Susarla Sitarama S Chakravarthy

James T Motter

Pranabh Kumar Thaduri

Kishore Tupili

Naveen Awasthy

Deepaganesh Deepaganesh Paulraj

Andrei Son

Bhanu Prakash Padiri

Tejas Ambekar

Methods and systems for managing use and access to memory installed within a data processing system are disclosed. Primary memory (e.g., RAM such as DRAM, SRAM, or the like) that is traditionally non-persistent (e.g., volatile) is configured by a management entity hosted within the data processing system to become persistent (e.g., pseudo-persistent/pseudo non-volatile) to create a persistent reserved primary memory. Data may be stored in the created persistent reserved primary memory to allow processes that previously have to access slower secondary storage to obtain and use the data to be completed faster. The creation of the persistent reserved primary memory may be completed before a full start up process of the data processing system is completed. The data stored in the persistent reserved primary memory may also be retained (e.g., may not be lost) across reboots of the data processing system.

Reserved persistent random access memory

Embodiments for a secure data migration service across on-premise and multi-cloud environment(s). are presented herein. A first service encrypts data using a first encryption key to obtain a first version of encrypted data; sends the first version of encrypted data to a second service; and receives, from the second service, a second version of encrypted data—the second version of encrypted data comprising the first version of encrypted data having been further encrypted with a second encryption key corresponding to the second service. The first service further decrypts the second version of encrypted data using the first encryption key to obtain a third version of encrypted data being encrypted with the second encryption key; and sends the third version of encrypted data to the second service to facilitate decryption, using the second encryption key, of the third version of encrypted data to obtain the data.

Secure data migration service across on-premise and multi-cloud environment(s)

In response to reception of a hypercall addition request, a virtual machine receives a public key. Based on the public key, the virtual machine determines whether the hypercall addition request is valid. In response to the hypercall being valid, the virtual machine adds a hypercall associated with the hypercall addition request within a hypervisor of an information handling system.

Para-virtualized drivers to prevent non-internet protocol hypervisor hypercall denial of service attacks

Virtual memory pooling, including identifying GPUs of respective IHSs, wherein each of the GPUs is associated with a respective internal memory allocation; partitioning, for each GPU, the internal memory allocation associated with the GPU into a first memory allocation and a second memory allocation; allocating, for each GPU, the first memory allocation of the internal memory allocation associated with the GPU as accessible only by the associated GPU; pooling, for each GPU, the second memory allocation of the internal memory allocation associated with the GPU to define a virtual memory pool, the virtual memory pool accessible by each GPU; processing, at a first GPU, a computational task, including: accessing the first memory allocation associated with the first GPU; determining that processing of the computational task exceeds a capacity of the first memory allocation of the first GPU and in response, requesting access to the virtual memory pool.

Virtual memory pooling

An information handling system detects installation of a hardware device that includes software, and transmits a request for a secure boot certificate associated with a hardware identifier of the hardware device and a software version of the software to a remote secure boot service. The secure boot service which maintains a centralized secure boot certificate store queries for the secure boot certificate based on the request and transmits the secure boot certificate to the information handling system. The secure boot service also performs a refinement mapping of the secure boot certificates in the store. Subsequent to receiving the secure boot certificate, the system provisions the secure boot certificate.

Software defined storage aware hardware accelerated NVME initiator service

Denial of dynamic host configuration protocol (DHCP) internet protocol (IP) address allocation to an unauthorized node via cross secure boot verification is presented herein. A system enables storage of a boot storage certificate in a boot data store of a client device, in which the boot storage certificate has been signed by a key exchange key (KEK) that has been stored in a network boot data store of a network switch device of a network; intercepts a DHCP discover packet that has been directed, by the client device, to a DHCP device of the system that assigns respective IP addresses to client devices to facilitate respective accesses, via the client devices using the respective IP addresses, of the network; and in response to determining that the secure boot storage certificate has been signed by the KEK, assigns an IP address of the respective IP addresses to the client device.

Denial of dynamic host configuration protocol internet protocol address allocation to an unauthorized node via cross secure boot verification

In some embodiments, apparatuses and methods are provided herein useful to validating migrated data. In some embodiments, there is provided a system for validating migrated data including a control circuit configured to migrate data from a first database platform to a second database platform and validate the migrated data. The control circuit configured to transmit a message indicating a mismatch in response to a determination that a first single aggregated hash value does not match with a second single aggregated hash value.

Systems and methods for data validation

Temporary firmware is provided as cloud services. Different temporary firmware containers are downloaded via a communications network. A light-weight operating system launches and executes the temporary firmware containers during a boot operation, POST operation, or other scheme. The temporary firmware containers thus detect and perhaps resolve POST errors. The light-weight operating system may also download a full-service/resource operating system. A second or subsequent boot operation may be performed, but control is ceded to the full-service/resource operating system. Multiple firmware tenants may thus be temporarily downloaded to a bare metal machine to support POST error detection activities. Advanced OS serviceability, diagnostics, and other containerized firmware may thus be quickly and simply launched without requiring the excessive time and difficulties of using the full-service/resource operating system.

Containerized firmware services

This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for variable rate tessellation. A graphics processor may receive data for geometry processing of a plurality of patches in a draw call. The graphics processor may reduce a tessellation factor of each of the plurality of patches based on a property of each of the plurality of patches. The reduced tessellation factor may correspond to a TRF. The property may correspond to a shading rate or a number of visible pixels. The graphics processor may apply the TRF for each of the plurality of patches. The graphics processor may render each of the plurality of patches based on the applied TRF for each of the plurality of patches.

Variable rate tessellation

A system includes a virtual machine to transmit an input/output request to a data storage system and a hypervisor configured to maintain a map of the virtual machine to a virtual disk, wherein the virtual disk is a slice of a persistent storage device. A virtual machine server is configured to maintain a map of the virtual disk to a start address and an end address and to update the input/output request with the start address, the end address, and a virtual disk identifier associated with the virtual machine. A processor determines whether the start address and the end address are valid, and if the start address and the end address are valid, then process the input/output request. The response is transmitted to the input/output request.

System and method for slice virtual disk encryption

An information handling system in an infrastructure-as-a-service (IaaS) system includes a memory for storing code and a processor to execute code to instantiate an IaaS NVMe orchestrator. The IaaS NVMe orchestrator establishes a first fabric link with a first target NVMe array of storage resources of the IaaS, establishes a second fabric link with a second target NVMe array of the storage resources, and establishes a first session on the storage resources. The first session includes a first namespace on the first target NVMe array and a second namespace on the second target NVMe array. The first and second namespaces are mirrored namespaces. The IaaS NVMe orchestrator further determines that the first fabric link has failed, and establishes a third fabric link with the first target NVMe array in response to determining that the first fabric link has failed.

Software defined storage aware hardware accelerated NVMe initiator service

An information handling system includes a provisioning server and a server. The server includes a baseboard management controller (BMC) to determine a first hardware inventory profile for the server. The BMC provides the first hardware inventory profile to the provisioning server. The BMC stores first signed provisioning configuration content that is based on a first ownership certificate for a first owner of the server. The BMC determines a second hardware inventory profile for the server, and provides the second hardware inventory profile to the provisioning server. The BMC stores second signed provisioning configuration content that is based on a second ownership certificate for a second owner of the server. In response to an expiration of the second ownership certificate, the BMC removes the second signed provisioning configuration content, compares a current hardware inventory profile to the first hardware inventory profile, and generates a report to indicate any hardware changes.

Overlay ownership certificates including hardware inventory profiles

An information handling system includes a provisioning server and a server. The server includes a baseboard management controller (BMC) that configures a first ownership certificate for the server, and provides it to the provisioning server. The first ownership certificate is associated with a first owner. The BMC receives a first signed provisioning configuration content, and stores the first signed provisioning configuration content in an encrypted memory. The BMC configures a second ownership certificate for the server, and provides it to the provisioning server. The second ownership certificate is associated with a second owner. The BMC receives a second signed provisioning configuration content, and stores the second signed provisioning configuration content on top of the first signed provisioning configuration content in the encrypted memory. In response to an expiration of the second ownership certificate, the BMC removes the first signed provisioning configuration content, and applies the second signed provisioning configuration content.

Information handling system with overlay ownership certificates for ownership chaining

This disclosure provides systems, devices, apparatus, and methods, including computer programs encoded on storage media, for configurable aprons for expanded binning. Aspects of the present disclosure include identifying one or more pixel tiles in at least one bin and determining edge information for each pixel tile of the one or more pixel tiles. The edge information may be associated with one or more pixels adjacent to each pixel tile. The present disclosure further describes determining whether at least one adjacent bin is visible based on the edge information for each pixel tile, where the at least one adjacent bin may be adjacent to the at least one bin.

Configurable apron support for expanded-binning

The present disclosure relates to methods and apparatus for graphics processing. The present disclosure can calculate a center-edge distance of a first pixel, the center-edge distance of the first pixel equal to a distance from a first pixel center to one or more edges of a first primitive. Additionally, the present disclosure can store the center-edge distance of the first pixel when the first primitive is visible in a scene. The present disclosure can also determine an amount of overlap between the first pixel and the first primitive. Further, the present disclosure can blend a color of the first pixel with a color of a second pixel based on at least one of the center-edge distance of the first pixel or the amount of overlap between the first pixel and the first primitive.

Methods and apparatus for edge compression anti-aliasing

Growing community of inventors

Bengaluru, India

Ankit Kumar Singh