Fault domains and optimization domains in converged infrastructure information handling systems
An infrastructure services manager accesses resource description information identifying a plurality of information handling resources and a management endpoint corresponding to each of the resources. Management information for each resource is retrieved from each management endpoint and an infrastructure of the system is determined based on the management information. Placement domains, including optimization domains and fault domains, within the infrastructure are discovered. A domain description document identifying each placement domain, its type, and the server resources included may be generated. The endpoints may be monitored and, when an infrastructure change is detected, placement domains may be re-discovered. The resource description information may be included in a resource description file comprising structured data indicative of the plurality of information handling resources, the management endpoint corresponding to each of the information handling resources and a REST compliant protocol for a discovery interface for each of the information handling systems.
1. A method of managing an information handling system, the method comprising:
accessing resource description information identifying:
a plurality of information handling resources included in the information handling system; and
a plurality of management endpoints including a management endpoint for each of the plurality of information handling resources;
retrieving management information for one or more of the information handling resources from one or more of the plurality of management endpoints;
based on the management information, determining an infrastructure of the information handling system;
discovering placement domains within the infrastructure, wherein said discovering includes at least one of:
identifying an optimization domain, comprising a group of one or more of the information handling resources within which each of a plurality of tiers of an application service may be instantiated such that inter-tier communication:
is internal to the optimization domain; and
satisfies a communication performance criterion; and
identifying a fault domain, comprising two or more independently available information handling resources;
generating a domain description document indicative of:
the placement domains, and, for each of the placement domains:
a domain type; and
one or more information handling resources included in the placement domain;
detecting an infrastructure change based on at least one of:
a resource status change indicated by one of the plurality of management endpoints; and
a domain description change, comprising a change in the domain description document; and
responsive to detecting the infrastructure change, re-discovering placement domains and re-generating the domain description document.
2. The method of claim 1 , wherein the domain description document comprises structured data.
3. The method of claim 1 , wherein accessing the resource description information includes accessing a resource description file comprising structured data indicative of:
the plurality of information handling resources;
the plurality of management endpoints; and
a representational state transfer (REST) compliant protocol for a discovery interface for each of the plurality of information handling resources.
4. The method of claim 3 , wherein retrieving management information from the plurality of management endpoints comprises:
invoking a protocol-specific adapter, corresponding to the REST compliant protocol, to:
provide credentials to and establish a connection with one or more of the plurality of management endpoints; and
request the management information from a management agent.
5. The method of claim 1 , wherein identifying the optimization domain comprises:
identifying a plurality of conceivable optimization domains, wherein the plurality of conceivable optimization domains depends, at least in part, on an environment in which the application service is deployed, wherein the environment is selected from: a physical deployment environment, a virtual deployment environment, and a containerized deployment environment; and
selecting a best-achievable optimization domain from the plurality of conceivable optimization domains in accordance with available information handling resources and one or more optimization domain constraints.
6. The method of claim 5 , wherein the one or more optimization domain constraints are selected from a group consisting of:
an symmetry constraint requiring symmetry between each of two or more instances of the application service instantiated on independently available resources within the fault domain; and
a level constraint requiring a particular fault domain level.
7. The method of claim 5 , wherein identifying the fault domain comprises:
identifying a plurality of conceivable fault domains, wherein the plurality of conceivable fault domains depends, at least in part, on the environment in which the application service is deployed; and
selecting a best-achievable optimization domain from the plurality of conceivable optimization domains in accordance with available information handling resources and one or more fault domain constraints.
8. The method of claim 7 , wherein the one or more fault domain constraints include a first fault domain constraint requiring the optimization domain having at least a threshold level of performance.
9. The method of claim 7 , wherein the plurality of conceivable optimization domains include optimization domains selected from:
physical optimization domains, including:
a single chassis, multiple module optimization domain, wherein the single chassis includes aggregated I/O switching;
a single chassis, multiple module optimization domain, wherein the single chassis includes pass-through I/O switching; and
a single rack, multiple server optimization domain comprising a plurality of rack mount servers on a common rack;
virtual optimization domains, including:
a single module optimization domain comprising a single modular server within a modular chassis;
a single server optimization domain comprising a single rack mount server;
a single chassis, multiple module optimization domain, wherein the single chassis includes aggregated I/O switching;
a single chassis, multiple module optimization domain, wherein the single chassis includes pass-through I/O switching; and
a single rack, multiple server optimization domain comprising a plurality of rack mount servers on a common rack; and
containerized optimization domains, including:
a single module optimization domain comprising a single modular server within a modular chassis;
a single server optimization domain comprising a single rack mount server;
a single chassis, multiple module optimization domain, wherein the single chassis includes aggregated I/O switching;
a single chassis, multiple module optimization domain, wherein the single chassis includes pass-through I/O switching; and
a single rack, multiple server optimization domain comprising a plurality of rack mount servers on a common rack.
10. The method of claim 7 , wherein the plurality of conceivable fault domains include fault domains selected from:
physical fault domains, including:
chassis level fault domains comprising independently operable resources on different modular chassis of a common rack;
rack level fault domains comprising independently operable resources on different racks within a data center; and
data center level fault domains comprising independently operable resources on different data centers;
fault domains for virtual deployments, including:
module level fault domains comprising independent operable resources on different modular servers of a single modular chassis;
chassis level fault domains comprising independently operable resources on different modular chassis of a common rack;
rack level fault domains comprising independently operable resources on different racks within a data center; and
data center level fault domains comprising independent data center level resources on different data centers.
11. An infrastructure services system comprising:
a processor; and
a non-transitory computer-readable medium, including processor-executable instructions that, when executed by the processor, cause the processor to perform operations comprising:
accessing a resource description file indicative of:
a plurality of information handling resources of an information handling system;
a plurality of management endpoints including one or more management endpoints corresponding to each of the plurality of information handling resources; and
a representational state transfer (REST) compliant protocol for a discovery interface for each of the plurality of information handling resources;
retrieving, from each of the plurality of management endpoints, management information for a corresponding information handling resource;
based on the management information, determining an infrastructure of the information handling system;
discovering placement domains within the infrastructure, wherein discovering the placement domains includes at least one of:
identifying an optimization domain, comprising a group of one or more of the information handling resources within which each of a plurality of tiers of an application service may be instantiated such that inter-tier communication:
is internal to the optimization domain; and
satisfies a communication performance criterion; and
identifying a fault domain, comprising two or more independently available information handling resources;
wherein retrieving management information from each management endpoint comprises:
invoking a protocol-specific adapter corresponding to the REST compliant protocol to:
provide credentials to the management endpoint to establish a connection with the management endpoint; and
request the management information from a management agent.
12. The information handling system of claim 11 , wherein the operations include:
generating a domain description document comprising structured data identifying:
the placement domains and, for each placement domain:
a domain type; and
one or more information handling resources included in the placement domain.
13. The information handling system of claim 12 , wherein the operations include:
detecting an infrastructure change based on at least one of:
a status change, for one of the plurality of information handling resources, indicated by an management endpoint; and
a domain description change, comprising a change in the domain description document; and
responsive to detecting the infrastructure change, re-discovering placement domains and re-generating the domain description document.
14. The information handling system of claim 11 , wherein the resource description file comprises structured data indicative of:
the plurality of information handling resources;
the plurality of management endpoints; and
the REST compliant protocol for the discovery interface for each of the plurality of information handling resources.
15. The information handling system of claim 11 , wherein identifying the optimization domain comprises:
identifying a plurality of conceivable optimization domains, wherein the plurality of conceivable optimization domains depends, at least in part, on an environment in which the application service is deployed, wherein the environment is selected from: a physical deployment environment, a virtual deployment environment, and a containerized deployment environment; and
selecting a best-achievable optimization domain from the plurality of conceivable optimization domains in accordance with available information handling resources and one or more optimization domain constraints.
16. The information handling system of claim 15 , wherein the one or more optimization domain constraints are selected from a group consisting of:
a symmetry constraint requiring symmetry between each of two or more instances of the application service instantiated on independently available resources within the fault domain; and
an optimization domain constraint requiring a particular fault domain level.
17. The information handling system of claim 15 , wherein identifying the fault domain comprises:
identifying a plurality of conceivable fault domains, wherein the plurality of conceivable fault domains depends, at least in part, on the environment in which the application service is deployed; and
selecting a best-achievable optimization domain from the plurality of conceivable optimization domains in accordance with available information handling resources and one or more fault domain constraints.
18. The information handling system of claim 17 , wherein fault domain constraints include a level constraint requiring the optimization domain to have at least a threshold level of performance.
19. The information handling system of claim 17 , wherein the plurality of conceivable optimization domains include optimization domains selected from:
optimization domains for physical deployments, including:
a single chassis, multiple module optimization domain, wherein the single chassis includes aggregated I/O switching;
a single chassis, multiple module optimization domain, wherein the single chassis includes pass-through I/O switching; and
a single rack, multiple server optimization domain comprising a plurality of rack mount servers on a common rack;
optimization domains for virtual deployments, including:
a single module optimization domain comprising a single modular server within a modular chassis;
a single server optimization domain comprising a single rack mount server;
a single chassis, multiple module optimization domain, wherein the single chassis includes aggregated I/O switching;
a single chassis, multiple module optimization domain, wherein the single chassis includes pass-through I/O switching; and
a single rack, multiple server optimization domain comprising a plurality of rack mount servers on a common rack; and
optimization domains for containerized deployments, including:
a single module optimization domain comprising a single modular server within a modular chassis;
a single server optimization domain comprising a single rack mount server;
a single chassis, multiple module optimization domain, wherein the single chassis includes aggregated I/O switching;
a single chassis, multiple module optimization domain, wherein the single chassis includes pass-through I/O switching; and
a single rack, multiple server optimization domain comprising a plurality of rack mount servers on a common rack.
20. The information handling system of claim 17 , wherein the plurality of conceivable fault domains include fault domains selected from:
fault domains for physical deployments, including:
chassis level fault domains comprising independently operable resources on different modular chassis of a common rack;
rack level fault domains comprising independently operable resources on different racks within a data center; and
data center level fault domains comprising independently operable resources on different data centers;
fault domains for virtual deployments, including:
module level fault domains comprising independent operable resources on different modular servers of a single modular chassis;
chassis level fault domains comprising independently operable resources on different modular chassis of a common rack;
rack level fault domains comprising independently operable resources on different racks within a data center; and
data center level fault domains comprising independent data center level resources on different data centers.