Abstract: Systems and techniques that facilitate automated validation of power topology are provided. In various embodiments, a control component can transmit a transition command to a power-distribution node of a data center, wherein the transition command can cause an outlet of the power-distribution node to transition between power states. In various aspects, a verification component can verify that a power-consumption node of the data center is connected to the outlet by comparing a pre-transition power characteristic of the power-consumption node with a post-transition power characteristic of the power-consumption node.

Abstract: An embodiment includes extracting statistical data associated with invocation of an application programming interface (API) from a log and using the statistical data to calculate a performance value and generate an aggregate dataset that combines the performance value with performance values associated with other invocations of the API. The embodiment includes calculating metric values for performance values for respective time intervals of a time period and calculating mean and standard deviation values of the metric values for the time period. The embodiment includes selecting the API as a candidate API and detecting a Customer Impacting Event (CIE) by applying a machine learning algorithm using monitored values associated with the candidate API during a time frame defined by a rolling window. The embodiment also includes automatically initiating a selected alert from among a plurality of alert options based at least in part on the monitored values associated with the CIE.

Abstract: A method for determining an intervention response category for provisioning workflows. The method determines provisioning features of a provisioning step. The method performs outlier detection to identify and remove outliers from non-intervention data to produce a non-intervention normal data set. The method performs iterative grouping on the non-intervention normal data set to determine significant variables in the provisioning features. The method performs response mapping of provisions using results of the iterative grouping including a significance of errors and a presence of errors in the non-intervention normal data and partial intervention data to categorize the provisions into a response category.

Abstract: A method for determining an intervention response category for provisioning workflows. The method determines provisioning features of a provisioning step. The method performs outlier detection to identify and remove outliers from non-intervention data to produce a non-intervention normal data set. The method performs iterative grouping on the non-intervention normal data set to determine significant variables in the provisioning features. The method performs response mapping of provisions using results of the iterative grouping including a significance of errors and a presence of errors in the non-intervention normal data and partial intervention data to categorize the provisions into a response category.

Abstract: A system is configured to determine a dominant error causing a provisioning step to become stuck during provisioning of a machine in a cloud environment. The system includes memory for storing instructions, and a processor configured to execute said instructions to determine an inverse error frequency (IEF) value for pre-intervention errors in a set of intervention provisioning data; determine a dominant error for a provision during said provisioning step in said set of intervention provisioning data based on a pre-intervention error that has a maximum IEF value; determine a duration frequency (DuF) value for the provision at said provisioning step for provisions in a set of non-intervention provisioning data; and determine said dominant error for each provision during said provisioning step in said set of non-intervention provisioning data based on an error that resulted in DuF value.

Abstract: A system is configured to determine a dominant error causing a provisioning step to become stuck during provisioning of a machine in a cloud environment. The system includes memory for storing instructions, and a processor configured to execute said instructions to determine an inverse error frequency (IEF) value for pre-intervention errors in a set of intervention provisioning data; determine a dominant error for a provision during said provisioning step in said set of intervention provisioning data based on a pre-intervention error that has a maximum IEF value; determine a duration frequency (DuF) value for the provision at said provisioning step for provisions in a set of non-intervention provisioning data; and determine said dominant error for each provision during said provisioning step in said set of non-intervention provisioning data based on an error that resulted in DuF value.

Abstract: An embodiment includes extracting statistical data associated with invocation of an application programming interface (API) from a log and using the statistical data to calculate a performance value and generate an aggregate dataset that combines the performance value with performance values associated with other invocations of the API. The embodiment includes calculating metric values for performance values for respective time intervals of a time period and calculating mean and standard deviation values of the metric values for the time period. The embodiment includes selecting the API as a candidate API and detecting a Customer Impacting Event (CIE) by applying a machine learning algorithm using monitored values associated with the candidate API during a time frame defined by a rolling window. The embodiment also includes automatically initiating a selected alert from among a plurality of alert options based at least in part on the monitored values associated with the CIE.

Abstract: Systems and techniques that facilitate automated validation of power topology are provided. In various embodiments, a control component can transmit a transition command to a power-distribution node of a data center, wherein the transition command can cause an outlet of the power-distribution node to transition between power states. In various aspects, a verification component can verify that a power-consumption node of the data center is connected to the outlet by comparing a pre-transition power characteristic of the power-consumption node with a post-transition power characteristic of the power-consumption node.

Abstract: A distributed power management system is configured determine a node power consumption of a node during a first time interval. The system can determine a node power cap. The system can determine a proportional component power budget for a component of the node based, at least in part, on the node power consumption and a component power consumption. The system can determine a power budget for the component for a second time interval based, at least in part on the proportional component power budget.

Abstract: A system and method for storing data that includes at least one memory device having a plurality of memory cells for storing data; and a memory control circuit that manages the read current and read pulse width applied to the memory cells, wherein the at least one memory device has a read current circuit configured to utilize adjustments to at least one of the read current or the read pulse width applied to the memory cells. In response to a request to read a group of the memory cells, the memory control circuit in an example, in response to determining that a comparative temperature value exceeds a first threshold, is configured to perform at least one of reducing the read current and/or increasing the read pulse width to be applied to the group of memory devices to be read.

Abstract: A computer controls power distribution. The computing system determines a power budget for a portion of a topography for a power delivery system. The computing system generates a pool of worker programs for the portion of the topography. The computing system generates a first number of power management tasks to manage power consumption in the portion of the topography based on the power budget. The computing system sends the first number of power management tasks to at least one worker program included in the pool of worker programs.

Abstract: Technology for computing number of active servers needed over time in a cloud/compute cluster includes the following operations (not necessarily in the following order): (i) determining the capacity of each VCE provisioned on the cloud against the resource guaranteed to that VCE; (ii) forecasting the resource needs over time using historical requests for each VCE flavor; and (iii) using the forecasted resource needs to determine the required number of future servers at some future time. Some embodiments of the present invention use a formula that accounts for the interplay among various parameter values of the VCE flavors and also the mapping of the needs of VCEs of various flavors to the capabilities of physical resources.

Abstract: Technology for computing number of active servers needed over time in a cloud/compute cluster includes the following operations (not necessarily in the following order): (i) determining the capacity of each VCE provisioned on the cloud against the resource guaranteed to that VCE; (ii) forecasting the resource needs over time using historical requests for each VCE flavor; and (iii) using the forecasted resource needs to determine the required number of future servers at some future time. Some embodiments of the present invention use a formula that accounts for the interplay among various parameter values of the VCE flavors and also the mapping of the needs of VCEs of various flavors to the capabilities of physical resources.

Abstract: A system and method for storing data that includes at least one memory device having a plurality of memory cells for storing data; and a memory control circuit that manages the read current and read pulse width applied to the memory cells, wherein the at least one memory device has a read current circuit configured to utilize adjustments to at least one of the read current or the read pulse width applied to the memory cells. In response to a request to read a group of the memory cells, the memory control circuit in an example, in response to determining that a comparative temperature value exceeds a first threshold, is configured to perform at least one of reducing the read current and/or increasing the read pulse width to be applied to the group of memory devices to be read.

Abstract: Memory bandwidth allocations for a plurality of tenants on a processor socket may be managed by determining a total memory bandwidth for the processor socket, determining a respective share of the total memory bandwidth for each tenant, and determining a respective dynamic memory bandwidth budget for the each tenant. The memory bandwidth allocations may be controlled based on the determined respective dynamic memory bandwidth budgets, wherein the each respective dynamic memory bandwidth budget supports utilizing the each respective share and supports sharing unused memory bandwidth of the each respective tenant with a different tenant included in the plurality of tenants.

Abstract: Dynamically allocating cache in a multi-tenant infrastructure includes monitoring cache usage for multiple workloads in a multi-tenant processing infrastructure to determine a workload phase. A baseline performance level per workload is determined. The baseline performance level is dependent upon the workload phase. The workloads for each tenant are categorized based on cache utilization and the cache is allocated to each workload based on the baseline performance level, cache utilization, and system wide cache capacity.

Abstract: Utilizing a computing device to determine and enforce limits on cloud computing containers receiving data over a network. A determination is made of total container time remaining available for a first container to execute in a computing environment. Processor packet receipt time is determined for receiving and processing of a packet or a batch of packets via a network stack associated with the computing device. An updated total container time remaining is calculated for the first container accounting for the processor packet receipt time. The updated total container time remaining is enforced by dropping a subsequent packet or batch of packets received at the network stack if the updated total container time remaining is insufficient.

Abstract: Technology for computing number of active servers needed over time in a cloud/compute cluster includes the following operations (not necessarily in the following order): (i) determining the capacity of each VCE provisioned on the cloud against the resource guaranteed to that VCE; (ii) forecasting the resource needs over time using historical requests for each VCE flavor; and (iii) using the forecasted resource needs to determine the required number of future servers at some future time. Some embodiments of the present invention use a formula that accounts for the interplay among various parameter values of the VCE flavors and also the mapping of the needs of VCEs of various flavors to the capabilities of physical resources.

Abstract: Technology for computing number of active servers needed over time in a cloud/compute cluster includes the following operations (not necessarily in the following order): (i) determining the capacity of each VCE provisioned on the cloud against the resource guaranteed to that VCE; (ii) forecasting the resource needs over time using historical requests for each VCE flavor; and (iii) using the forecasted resource needs to determine the required number of future servers at some future time. Some embodiments of the present invention use a formula that accounts for the interplay among various parameter values of the VCE flavors and also the mapping of the needs of VCEs of various flavors to the capabilities of physical resources.

Abstract: A computer controls power distribution. The computing system determines a power budget for a portion of a topography for a power delivery system. The computing system generates a pool of worker programs for the portion of the topography. The computing system generates a first number of power management tasks to manage power consumption in the portion of the topography based on the power budget. The computing system sends the first number of power management tasks to at least one worker program included in the pool of worker programs.