Technologies for dividing work across accelerator devices
US-2024143410-A1 · May 2, 2024 · US
US9021076B2 · US · B2
| Field | Value |
|---|---|
| Publication number | US-9021076-B2 |
| Application number | US-201113177704-A |
| Country | US |
| Kind code | B2 |
| Filing date | Jul 7, 2011 |
| Priority date | Jul 7, 2011 |
| Publication date | Apr 28, 2015 |
| Grant date | Apr 28, 2015 |
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.
Techniques are disclosed for managing inventory data for components of a server system. In one embodiment, a global management controller is provided, that is operatively connected to a plurality of local management controllers. Each local management controller is configured to manage a subset of the components of the server system. Each local management controller is also configured to generate, for each component, a checksum based on vital product data (VPD) of the component. Each local management controller is also configured to compute a composite checksum based on the checksums generated for the components in the subset. The global management controller is configured to maintain a global view of the VPD in the computer system, based on the checksums and/or composite checksums.
Opening claim text (preview).
What is claimed is: 1. A system, comprising: a global management controller having one or more computer processors configured to perform an operation for managing vital product data (VPD) associated with a plurality of components, each component configured to provide the associated VPD responsive to a request; and a plurality of local management controllers operatively connected to the global management controller via a network, each local management controller configured to manage a respective subset of the plurality of components; wherein the operation comprises: receiving, from a first local management controller of the plurality of local management controllers: (i) respective VPD of first and second components in the subset managed by the first local management controller; (ii) respective component checksums generated based on the VPDs of the first and second components; and (iii) a composite checksum specific to the first local management controller and generated based on the component checksums; storing the VPDs, component checksums, and composite checksum by the global management controller; receiving, from the first local management controller, a current composite checksum specific to the first local management controller; only upon determining that the stored composite checksum is non-current, retrieving current component checksums of the first and second components, respectively; comparing the retrieved current component checksums to the respective stored component checksum in order to identify at least one non-current stored component checksum; and then only for the at least one identified non-current stored component checksum, retrieving a current VPD of the respective component in order to update the stored VPD of the respective component, thereby reducing an amount of network traffic transmitted over the network. 2. The system of claim 1 , wherein the operation further comprises, upon determining that the stored VPD associated with the subset is non-current: identifying at least one component in the subset managed by the second local management controller for which VPD stored by the global management controller is non-current; and updating the stored VPD of the identified component using VPD retrieved from the second local management controller. 3. The system of claim 2 , wherein the at least one component is identified by comparing a stored checksum of the at least one component with a subsequently retrieved checksum of the at least one component. 4. The system of claim 1 , wherein each component is uniquely identifiable by an associated identifier, wherein each identifier is immutable, and wherein each checksum is generated via a cyclic redundancy check (CRC) algorithm. 5. The system of claim 1 , wherein the vital product data comprises at least one of a serial number, a part number, a version number, and a universally unique identifier (UUID). 6. A global management controller configured to manage component vital product data (VPD) and operatively connected, via a network, to a plurality of local management controllers, each local management controller configured to manage a respective subset of a plurality of components, each component having associated VPD, the global management controller comprising: a computer processor; and a memory storing management firmware which, when executed on the computer processor, performs an operation comprising: receiving, from a first local management controller of the plurality of local management controllers: (i) respective VPD of first and second components in the subset managed by the first local management controller, (ii) respective component checksums generated based on the VPDs of the first and second components; and (iii) a composite checksum specific to the first local management controller and generated based on the component checksums; storing the VPDs, component checksums, and composite checksum by the global management controller; receiving, from the first local management controller, a current composite checksum specific to the first local management controller; only upon determining that the stored composite checksum is non-current, retrieving current component checksums of the first and second components, respectively; comparing the retrieved current component checksums to the respective stored component checksum in order to identify at least one non-current stored component checksum; and then only for the at least one identified non-current stored component checksum, retrieving a current VPD of the respective component in order to update the stored VPD of the respective component, thereby reducing an amount of network traffic transmitted over the network. 7. The global management controller of claim 6 , wherein the operation further comprises, upon determining that the stored VPD associated with the subset is non-current: identifying at least one component in the subset managed by the second local management controller, for which VPD stored by the global management controller is non-current; and updating the stored VPD of the identified at least one component using VPD retrieved from the second local management controller. 8. The global management controller of claim 7 , wherein the at least one component is identified by comparing a stored checksum of the at least one component with a subsequently retrieved checksum of the at least one component. 9. The global management controller of claim 6 , wherein each component is uniquely identifiable by an associated identifier, wherein each identifier is immutable, and wherein each checksum is generated via a cyclic redundancy check (CRC) algorithm. 10. The global management controller of claim 6 , wherein the vital product data comprises at least one of a serial number, a part number, a version number, and a universally unique identifier (UUID). 11. The system of claim 1 , wherein the at least one component is identified by comparing a stored checksum of the at least one component with a subsequently retrieved checksum of the at least one component, wherein the current composite checksum specific to the first local management controller is received subsequent to a restart of the global management controller, wherein the amount of network traffic transmitted over the network subsequent to the restart of the global management controller is reduced, wherein the operation further comprises, upon determining that the stored VPD associated with the subset is non-current: identifying at least one component in the subset managed by the second local management controller for which VPD stored by the global management controller is non-current; and updating the stored VPD of the identified component using VPD retrieved from the second local management controller. 12. The system of claim 11 , wherein the operation further comprises: determining that the stored VPD associated with the subset managed by the second local management controller is current, whereupon no VPD associated with the subset managed by the second local management controller is retrieved by the global management controller from the second local management controller, thereby further reducing the amount of network traffic transmitted over the network subsequent to the restart of the global management controller; wherein upon determining that the stored VPD associated with the subset managed by the second local management controller is current, none of the plurality of checksums is retrieved by the global management controller from the second local management controller, thereby still further reducing the amount of network traffic transmitted over the network subsequent to the restart o
Computing infrastructure, e.g. computer clusters, blade chassis or hardware partitioning (casings, cabinets, racks or drawers for data centers H05K5/00) · CPC title
Electricity · mapped topic
for remote control or remote monitoring of applications · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.