Amazon Technologies, Inc.

Nvidia Corporation

Rahul Gautam Patel

Nachiketh Rao Potlapally

William John Earl

Derek Del Miller

Matthew Shawn Wilson

Mark Bradley Davis

Eric Jason Brandwine

Anthony Nicholas Liguori

Asif Khan

David James Borland

Gary D Hicok

Uwe Dannowski

Matthew Raymond Longnecker

A multi-tenant trusted platform module (MTTPM) is attached to a communication bus of a virtualization host. The MTTPM includes a plurality of per-guest-virtual-machine (per-GVM) memory location sets. In response to an indication of a first trusted computing request (TCR) associated with a first GVM of a plurality of GVMs instantiated at the virtualization host, a first memory location of a first per-GVM memory location set is accessed to generate a first response indicative of a configuration of the first GVM. In response to an indication of a second TCR associated with a second GVM, a second memory location of a second-per-GVM memory location set is accessed to generate a second response, wherein the second response is indicative of a different configuration of the second GVM.

Multi-tenant trusted platform modules

Techniques are described for injecting noise in a timer value provided to an instruction requesting the timer value. A plurality of tasks may execute on a processor, wherein the processor may comprise one or more processing cores and each task may include a plurality of computer executable instructions. In accordance with one technique for injecting noise in the timer value, a request for a first timer value is received by one or more computer executable instructions belonging to a first task from the plurality of tasks, and in response, a second timer value is provided to the first task instead of the first timer value, wherein the second timer value is derived from the first timer value and a random number.

Noise injected virtual timer

A computing device includes a processor and a persistent memory for storing information about a first public key associated with a first asymmetric key pair for authenticating the source of a digital certificate. The computing device also includes a second memory for storing one or more current key version indicators. Each of the current key version indicators is associated with a corresponding secondary public key, and the one or more current key version indicators are used by the processor to determine the trust of the corresponding secondary public key.

Controlling use of encryption keys

Techniques are described for accumulating unused computing resources. The techniques may involve associating, with a task, a baseline amount of a computing resource for each time interval of a predetermined number of time intervals, and monitoring a consumption of the computing resource by the task in each time interval. Resource credits can be accumulated based on an unused amount of the computing resource during at least some of the time intervals. When a workload of the task consumes more than the baseline amount of the computing resource, the accumulated resource credits can be applied to allocate an additional amount of the computing resource to the task. A proportionate additional amount of a memory resource can also be allocated to the task.

Allocating processor resources based on a task identifier

A computing device includes a processor and a persistent memory for storing information about a first public key associated with a first asymmetric key pair for authenticating the source of a digital certificate. The computing device also includes a second memory for storing one or more current certificate version indicators, each associated with a corresponding digital certificate, and the version indicator is used by the processor to determine the trust of the corresponding digital certificate.

Controlling digital certificate use

Techniques are described for allocating resources to a task from a shared hardware structure. A plurality of tasks may execute on a processor, wherein the processor may include one or more processing cores and each task may include a plurality of computer executable instructions. In accordance with one technique for allocating resources to a task from a shared hardware structure amongst multiple tasks, aspects of the disclosure describe assigning a first identifier to a first task from the plurality of tasks, associating a portion of the shared hardware resource with the first identifier, and restricting access and/or observability for computer executable instructions executed from any other task than the first task to the portion of the hardware resource associated with the first identifier.

Allocation of shared system resources

Techniques are described for allocating computing resources to a task from a shared hardware structure. The techniques may involve receiving a request to execute a task for a tenant on shared hardware resources, and determining a set of computing resources for allocation to the task based on a service level agreement associated with the tenant. The set of computing resources can be allocated to the task based on the service level agreement associated with the tenant. In some aspects, one or more performance counters associated with one or more of the computing resources can be monitored to determine an activity level for the one or more computing resources during execution of the task, and one or more allocations of the computing resources for execution of the task can be adjusted based on the activity level for the one or more computing resources.

Allocating processor resources based on a service-level agreement

One or more resources for an SoC can be directly mapped to a host address space in a host system as peripheral bus functions. A translation unit can provide translation between the host address space and an SoC address space for transactions targeted for a resource from the one or more resources to facilitate performing the transactions with the resource using the host address space. Some embodiments of the technology can provide peer to peer capability for communication between the SoC resources using the translation unit.

Resource mapping for an input/output device

The performing of virtual machine (VM)-based secure operations is enabled using a trusted co-processor that is able to operate in a secure mode to perform operations in a multi-tenant environment that are protected from other VMs and DOM-0, among other domains and components. A customer VM can contact a VM manager (VMM) to perform an operation with respect to sensitive data. The VMM can trigger secure mode operation, whereby memory pages are marked and access blocked to entities outside a trusted enclave. The trusted co-processer can measure the VMM and compare the result against an earlier result to ensure that the VMM has not been compromised. Once the operations are performed, the trusted co-processor can return the results, and the VMM can exit the secure mode such that access to the marked pages and customer data is restored.

Platform secure execution modes

Techniques are described for injecting noise in a timer value provided to an instruction requesting the timer value. A plurality of tasks may execute on a processor, wherein the processor may comprise one or more processing cores and each task may include a plurality of computer executable instructions. In accordance with one technique for injecting noise in the timer value, in response to a request for a timer value, an artificial and indeterminate amount of delay may be introduced before accessing of the timer value from the hardware timer. In one implementation, access to the hardware timer for the timer value may be gated by one or more artificially injected micro-architectural events.

Micro-architecturally delayed timer

A technique for managing processor cores within a multi-core central processing unit (CPU) provides efficient power and resource utilization over a wide workload range. The CPU comprises at least one core designed for low power operation and at least one core designed for high performance operation. For low workloads, the low power core executes the workload. For certain higher workloads, the high performance core executes the workload. For certain other workloads, the low power core and the high performance core both share execution of the workload. This technique advantageously enables efficient processing over a wider range of workloads than conventional systems.

Heterogeneous multiprocessor design for power-efficient and area-efficient computing

Techniques are described for accumulating unused computing resources. The techniques may involve receiving a request to execute a task for a tenant on shared hardware resources and assigning a task identifier to the task. A baseline resource credit per time interval of a computing resource can be determined based on a service level agreement associated with the tenant. The techniques may further involve monitoring a performance counter associated with the computing resource to determine a utilization amount of the computing resource by the task during a first time interval, determining an unused amount of the computing resource in the first time interval based on a difference between the utilization amount and the baseline resource credit and incrementing a resource credit balance associated with the task identifier by the unused amount of the computing resource.

Growing community of inventors

Austin, TX, United States of America

Rahul Gautam Patel