Automated discovery and inventory of nodes within an autonomic distributed computing system

• US 7,590,653 B2
• Filed: 03/02/2005
• Issued: 09/15/2009
• Est. Priority Date: 03/02/2005
• Status: Expired due to Fees

• Current Assignee: Computer Associates Think Inc. (Broadcom, Inc.)
• Original Assignee: Cassatt (Broadcom, Inc.)
• Inventors: Sparks, Jonathan
• Primary Examiner(s): Vital; Pierre M
• Assistant Examiner(s): Nofal; Christopher P
• Application Number:
  US11/070,851
• Publication Number:
  US 20060200494A1
• Time in Patent Office: 1,658 Days
• Field of Search: 707/10, 707/102, 707/104.1, 709/221, 709/222, 709/223, 709/224, 709/226
• US Class Current: 1/1
• CPC Class Codes: H04L 41/0806 for initial configuration or provisioning, e.g. plug-and-play
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 41/14 Network analysis or design
• CPC Class Codes: H04L 61/50 Address allocation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99946 Object-oriented database structure network

Abstract

A distributed computing system conforms to a multi-level, hierarchical organizational model. One or more control nodes provide for the efficient and automated allocation and management of computing functions and resources within the distributed computing system in accordance with the organization model. The model includes four distinct levels: fabric, domains, tiers and nodes that provide for the logical abstraction and containment of the physical components as well as system and service application software of the enterprise. A user, such as a system administrator, interacts with the control nodes to logically define the hierarchical organization of distributed computing system. The control node detects the addition of a node added to the network and automatically identifies attributes for the detected node.

Claims

1. A computer-implemented method comprising:

storing a model within a database of a control node, wherein the model defines an organization of a distributed computing system having a plurality of computing nodes;

automatically detecting the addition of a node to a network;

upon detecting the addition of the node, network booting the detected node with an inventory software image that primarily only includes an executable inventory process for identifying hardware attributes of the detected node;

after network booting the detected node, executing the inventory process on the detected node to automatically identify the hardware attributes for the detected node;

sending the identified hardware attributes from the detected node to the control node for storage within the model;

updating the model to store the identified hardware attributes for the detected node;

automatically deploying the detected node within the distributed computing system in accordance with the model based on the identified hardware attributes;

receiving an input that defines the model in a hierarchical form that includes a fabric having one or more domains, wherein each domain has at least one tier that performs a computing function, wherein each tier includes at least one node slot; and

automatically configuring he distributed computing system in accordance with the input, wherein automatically configuring the distributed computing system comprises;

comparing the identified hardware attributes of one or more nodes against one or more tier requirements;

automatically assigning the nodes to the node slots of the tiers based on the identified hardware attributes, the tier requirements, and a minimum positive processing energy for at least one node;

automatically mapping software images to the node slots in accordance with the computing functions of the tiers; and

automatically loading the software images onto the nodes to provide a computing environment for execution of user software applications, wherein automatically loading the software images comprises automatically loading a software image onto the detected node to replace the inventory software image.

2. The method of claim 1, wherein automatically detecting the addition of the node comprises:

detecting automatic assignment of a temporary network address to the node; and

analyzing the assignment to determine whether the node is a candidate for deployment as one of the computing nodes within the distributed computing system.

3. The method of claim 2, wherein analyzing the assignment comprises:

parsing the assignment of the temporary network address to extract an identifier; and

determining whether the identifier matches one of a set of known types of computing nodes.

4. The method of claim 1, further comprising:

creating a node object within the model upon detecting the node; and

updating the node object to store the attributes.

5. The method of claim 1, wherein automatically detecting the addition of the node comprises:

determining a media access controller (MAC) address for the detected node; and

storing the MAC address within a node object of the model.

6. The method of claim 5, wherein determining a MAC address comprises:

remotely cycling power on the detected node;

remotely monitoring a console port of the detected node to capture textual configuration information while the detected node reboots; and

parsing the textual configuration to extract the MAC address.

7. The method of claim 1, wherein automatically detecting the addition of the node comprises:

determining whether the detected node is a control node for a plurality of nodes;

creating a respective node object within the model for each of the plurality of the nodes;

automatically identifying attributes for each of the plurality of nodes; and

updating the node objects within the model to store the attributes.

8. The method of claim 1, further comprising:

automatically updating the model to assign the detected node to a discovered pool upon detecting the node; and

automatically updating the model to assign the detected node to a free pool upon identifying the attributes.

9. The method of claim 1, wherein network booting the detected node comprises:

configuring a pre-boot execution environment to define the software image to include the inventory process, an operating system kernel, a random access memory (RAM) disk, and a network interface.

10. The method of claim 1, further comprising:

executing the inventory process on the detected node to identify a plurality of network interfaces on the detected node;

determining which of the network interfaces can communicate with each other; and

updating the model to mark the interfaces that can communicate with each other as bonded interfaces.

11. The method of claim 1, further comprising:

executing the inventory process on the detected node to generate inventory data that conforms to a data description language.

12. The method of claim 1, wherein the inventory software image further comprises an operating system kernel, a random access memory (RAM) disk, and a network interface.

13. The method of claim 1, wherein the software image is a complete set of software loaded on the node including an operating system and all boot code, middleware and application files necessary to provide the computing environment for execution of the user software applications.

14. A distributed computing system comprising:

a plurality of application nodes interconnected via a network;

a database that stores a model that defines an organization for the distributed computing system; and

a control node that detects the addition of a node to the network and automatically identifies hardware attributes for the detected node, wherein the control node comprises an inventory service that network boots the detected node with an inventory software image when the control node detects the addition of the node, wherein the inventory software image primarily only includes an executable inventory process for identifying hardware attributes of the detected node, wherein, after network booting the detected node, the inventory process executes on the detected node to automatically identify the hardware attributes for the detected node and sends the identified hardware attributes from the detected node to the control node for storage within the model, wherein the control node updates the model to store the identified hardware attributes for the detected node, automatically deploys the detected node within the distributed computing system in accordance with the model based on the identified hardware attributes, receives an input that defines the model in a hierarchical form that includes a fabric having one or more domains, wherein each domain has at least one tier that performs a computing function, wherein each tier includes at least one node slot, and wherein the control node automatically configures the distributed computing system in accordance with the input, compares the identified hardware attributes of one or more nodes against one or more tier requirements, automatically assigns the nodes to the node slots of the tiers based on the identified hardware attributes, the tier requirements, and a minimum positive processing energy for at least one node, automatically maps software images to the node slots in accordance with the computing functions of the tiers, automatically loads the software images onto the nodes to provide a computing environment for execution of user software applications, and automatically loads a software image onto the detected node to replace the inventory software image.

15. The distributed computing system of claim 14, further comprising:

a dynamic host communication protocol (DHCP) server that assigns a temporary network address to the node,

wherein the control node further comprises a discovery service that detects the assignment of the temporary network address and analyzes the assignment to determine whether the node is a candidate for a computing node to be deployed within the distributed computing system.

16. The distributed computing system of claim 15, wherein the discovery service parses the assignment made by the DHCP server to extract an identifier and determines whether the identifier matches one of a set of known types of computing nodes.

17. The distributed computing system of claim 15, wherein the discovery service creates a node object within the model upon detecting the node, and updates the node object to store the attributes.

18. The distributed computing system of claim 15,

wherein the discovery service executing on the control node remotely cycles power on the detected node and monitors a console port of the detected node to capture textual configuration information when the detected node reboots, and

wherein the discovery service parses the textual configuration to extract a media access control (MAC) address and stores the MAC address within the node object of the model.

19. The distributed computing system of claim 15,

wherein the discovery service determines whether the detected node is a control node for a plurality of nodes and creates a respective node object within the model for each of the plurality of the nodes, and

wherein the discovery service automatically identifies attributes for each of the plurality of nodes and updates the node objects to store the attributes.

20. The distributed computing system of claim 14,

wherein the control node further comprises a discovery service that automatically detects the node and updates the model to add the node to a pool of discovered nodes,

wherein the inventory service automatically identifies the attributes and updates the model to assign the node from the pool of discovered nodes to a pool of free nodes available for deployment within the distributed computing system.

21. The distributed computing system of claim 14, wherein the inventory service configures a pre-boot execution environment to define the software image to include the inventory process, an operating system kernel, a random access memory (RAM) disk, and a network interface.

22. The distributed computing system of claim 14,

wherein the inventory process executes on the detected node to identify a plurality of network interfaces on the detected node, and

wherein the inventory process determines which of the network interfaces can communicate with each other and marks the interfaces that can communicate with each other as bonded interfaces.

23. The distributed computing system of claim 14, wherein the inventory process executes on the detected node to generate inventory data that conforms to a data description language.

24. The method of claim 14, wherein the inventory software image further comprises an operating system kernel, a random access memory (RAM) disk, and a network interface.

25. The distributed computing system of claim 14, wherein the software image is a complete set of software loaded on the node including an operating system and all boot code, middleware and application files necessary to provide the computing environment for execution of the user software applications.

26. A computer-readable medium comprising instructions that cause a processor to:

store a model within a database of a control node, wherein the model defines an organization of a distributed computing system having a plurality of computing nodes;

automatically detect the addition of a node to a network;

network boot the detected node with an inventory software image upon detecting the addition of the node, wherein the inventory software image primarily only includes an executable inventory process for identifying hardware attributes of the detected node;

execute the inventory process on the detected node, after network booting the detected node, to automatically identify the hardware attributes for the detected node;

send the identified hardware attributes from the detected node to the control node for storage within the model;

update the model to store the identified hardware attributes for the detected node;

automatically deploy the detected node within the distributed computing system in accordance with the model based on the identified hardware attributes;

receive an input that defines the model in a hierarchical form that includes a fabric having-one or more domains, wherein each domain has at least one tier that performs a computing function, wherein each tier includes at least one node slot; and

automatically configure the distributed computing system in accordance with the input, wherein the instructions that cause the processor to automatically configure the distributed computing system comprise instructions that cause the processor to;

compare the identified hardware attributes of one or more nodes against one or more tier requirements;

automatically assign the nodes to the node slots of the tiers based on the identified hardware attributes, the tier requirements, and a minimum positive processing energy for at least one node;

automatically map software images to the node slots in accordance with the computing functions of the tiers; and

automatically load the software images onto the nodes to provide a computing environment for execution of user software applications, wherein the instructions that cause the processor to automatically load the software images comprise instructions that cause the processor to automatically load a software image onto the detected node to replace the inventory software image.

27. The computer-readable medium of claim 26, wherein the inventory software image further comprises an operating system kernel, a random access memory (RAM) disk, and a network interface.

28. The computer-readable medium of claim 26, wherein the software image is a complete set of software loaded on the node including an operating system and all boot code, middleware and application files necessary to provide the computing environment for execution of the user software applications.