Infrastructure driven auto-scaling of workloads
US-2024419470-A1 · Dec 19, 2024 · US
US9766948B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9766948-B2 |
| Application number | US-201514948277-A |
| Country | US |
| Kind code | B2 |
| Filing date | Nov 21, 2015 |
| Priority date | Nov 11, 2015 |
| Publication date | Sep 19, 2017 |
| Grant date | Sep 19, 2017 |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
An approach is provided to dynamically select a micro-threading (MT) mode of each core of a processor based on a load on each of the respective cores while the processor is running a hypervisor. The approach sets a core's micro-threading mode to a whole-core mode (MT 1 ) in response to identifying that the load on the selected core is at a light load level, sets the core's micro-threading mode to a two-way micro-threading mode (MT 2 ) in response to identifying that the load on the selected core has increased above the light load level, and sets the selected core's micro-threading mode to a four-way micro-threading mode (MT 4 ) in response to identifying that the load on the selected core is at a high load level.
Opening claim text (preview).
What is claimed is: 1. A method that dynamically selects a micro-threading (MT) mode of each core of a processor based on a load on each of the respective cores while the processor is running a hypervisor, the method comprising: dynamically setting a selected core's micro-threading mode to a whole-core mode (MT 1 ) in response to identifying that the load on the selected core is at a first load level; dynamically setting the selected core's micro-threading mode to a two-way micro-threading mode (MT 2 ) in response to identifying that the load on the selected core has increased above the first load level to a second load level; dynamically setting the selected core's micro-threading mode to a four-way micro-threading mode (MT 4 ) in response to identifying that the load on the selected core has increased above the second load level to a third load level; entering a quest context after the setting of the micro-threading mode; and switching the micro-threading mode of the selected core to the whole-core mode (MT 1 ) upon the selected core exiting from the guest context. 2. The method of claim 1 further comprising: selecting a primary virtual core to execute on the selected core; scanning a plurality of preempted virtual cores; and selecting one or more of the preempted virtual cores to execute as secondary virtual cores along with the primary virtual core. 3. The method of claim 2 further comprising: selecting the micro-threading mode based on the selection of the secondary virtual cores. 4. The method of claim 3 further comprising: in response to a task executing on the primary virtual core requiring a service from a host process: placing the task on a preempted virtual cores list corresponding to a current CPU; and selecting a different task to execute on the primary virtual core. 5. The method of claim 4 further comprising: performing the service by the host process on the task; and removing the task from the preempted virtual cores list. 6. The method of claim 3 further comprising: performing any needed service in a host process in response to exiting from the guest context; and for each virtual CPU (vCPU) task scheduled to execute on the selected virtual core: add an entry to a vCore data structure pertaining to each vCPU task that is ready to execute on the selected virtual core; selecting one of the vCPU tasks as the runner task for the selected virtual core, wherein the runner task executes on the primary virtual core; and selecting one or more of the vCPU tasks as tasks to execute concurrently with the runner task on the selected virtual core.
Hypervisor-specific management and integration aspects · CPC title
Techniques for rebalancing the load in a distributed system · CPC title
Starting, stopping, suspending or resuming virtual machine instances · CPC title
from multiple instruction streams, e.g. multistreaming · CPC title
according to execution mode, e.g. mode flag · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.