Virtual data center for network resource management

• US 20070027973A1
• Filed: 07/29/2005
• Published: 02/01/2007
• Est. Priority Date: 07/29/2005
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Scalent Systems, Inc. (Dell Technologies Inc.)
• Inventors: Bigall, Evan, Stein, Lawrence, Jaszewski, Gary, Oberhaus, Jared, Greer, Russell
• Granted Patent: US 7,574,491 B2
• Time in Patent Office: Days
• US Class Current: 709/223
• CPC Class Codes: H04L 41/046 comprising network management agents or mobile agents therefor
• CPC Class Codes: H04L 41/0806 for initial configuration or provisioning, e.g. plug-and-play
• CPC Class Codes: H04L 41/0869 Validating the configuration within one network element
• CPC Class Codes: H04L 41/12 Discovery or management of network topologies
• CPC Class Codes: H04L 41/40 using virtualisation of network functions or resources, e.g. SDN or NFV entities

Abstract

A system and a method for implementing a virtual data center. In which, the functions to be performed by a computer, and the access the computer equipment is required to have to the storage and network facilities, are decoupled from the hardware, and instantiated dynamically on any suitable computer equipment, without any manual intervention. In a virtual data center, all the physical resources are virtualized and arranged dynamically, to meet the functional requirements of the data center.

Claims

1. A virtual data center system for managing a network of servers, each server in the network of servers requiring a server image, the server image enabling booting of the server, the server image being stored over a storage system, the virtual data center system comprising:

a. a layer 2 interconnect fabric, the layer 2 interconnect fabric comprising a plurality of switches, each of the plurality of switches comprising a plurality of ports, the plurality of ports providing connectivity to the network of servers;

b. a controller, the controller dynamically configuring and assigning a server image, the server image being assigned to a server, the server making a request for the server image;

c. an agent, the agent residing on each server of the network of servers, the agent further enabling communication between the controller and the virtual data center system; and

d. a system management infrastructure, the system management infrastructure comprising remote management interfaces, the remote management interfaces providing a communication channel between each servers of the network of servers and the controller.

2. The virtual data center system of claim 1 further comprising a console, the console providing an interface to a data center administrator to communicate with the controller.

3. The virtual data center system of claim 2, wherein the console further includes a graphical tool, the graphical tool enabling the data center administrator to specify the overall functional design of the virtual data center system.

4. The virtual data center system of claim 1 further comprising a plurality of appliances, the appliances being connected to the layer 2 interconnect fabric.

5. The virtual data center system of claim 4, wherein the controller comprises a proxy agent, the proxy agent providing communication between the controller and the plurality of appliances.

6. The virtual data center system of claim 1 further comprising a system management network, the system management network connecting the controller to each server in the network of servers, the plurality of appliances, and the plurality of switches of the layer 2 interconnect fabric either directly or indirectly through management devices.

7. The virtual data center system of claim 1 further comprising a boot network, the boot network connecting each server in the network of servers to the layer 2 interconnect fabric, each server in the network of servers waiting for a server image to be assigned.

8. The virtual data center system of claim 1 further comprising a system control network, the system control network further providing communication between the agent and the controller when the corresponding server is in the booted state.

9. The virtual data center system of claim 1 further comprising an abstraction layer, the abstraction layer providing an OS (Operating System)-independent communication among the servers in the network of servers.

10. The virtual data center system of claim 1 further comprising a virtual rack, the virtual rack arranging a plurality of servers in groups, each of the plurality of servers in a group being connected to the layer 2 interconnect fabric through the ports on the same switch.

11. The virtual data center system of claim 1, wherein a server in the network of servers is selected from a group consisting of stand-alone servers, blade servers and virtual servers.

12. The virtual data center system of claim 1, wherein the storage system is selected from a group consisting of a local disk, a Storage Area Network (SAN), and a Network Attached Storage (NAS) device.

13. The virtual data center system of claim 1, wherein the controller further includes a controller database, the controller database providing storage facilities to store the abstracted server image.

14. The virtual data center of claim 13, wherein the abstracted server image comprises hardware attributes associated with each server image, network related attributes associated with each server image and the attributes associated with the access of storage systems.

15. A method for managing a virtual data center system, the virtual data center system comprising a plurality of servers, the plurality of servers being connected to a layer 2 interconnect fabric, the layer 2 interconnect fabric comprising a plurality of switches, each of the plurality of switches comprising a plurality of ports, each of the plurality of servers further having a server image, the sever image being required to boot the server, the server image being stored on a storage system, the virtual data center system further comprising a controller, the controller comprising a controller database, the controller database providing storage facility, the method comprising the steps of:

a. configuring the server images, the server images being configured dynamically by the controller, the server images being requested by the plurality of servers for booting;

b. allocating a server image to a server, wherein the server image is allocated by the controller; and

c. booting the server, wherein the booting of server is done with the allocated server image corresponding to the server.

16. The method of claim 15 further comprising the step of tracking the movement of the plurality of servers across the ports of the layer 2 interconnect fabric.

17. The method of claim 15 further comprising the step of providing a substitute server, the substitute server being provided when at least one of server impairment and server failure is reported, the server impairment being reported when a server fails to perform its functions and reports erroneous operational state, the server failure being reported when a server fails to report its status periodically.

18. The method of claim 15, wherein configuring the server images comprises the steps of:

a. abstracting the server images, wherein the hardware attributes associated with the server images, network related attributes and the attributes associated with the access of storage system are abstracted;

b. storing the abstracted server images in the controller database; and

c. assigning a unique server image identifier to each of the server images.

19. The method of claim 15, wherein allocating a server image to a server includes the steps of:

a. waiting for a server to be available or for a server image to make a request, the server being available when it does not have any assigned server image, the request being made by the server image for being allocated to a server;

if a server is available or if a server image requires allocation, performing step b;

b. checking whether a server image requires allocation;

if a server image requires allocation, performing step c;

c. checking whether the server image is locked for a particular server;

if the server image is locked for a particular server, performing step d;

d. booting the server with the locked server image;

else,

if the server image requires booting from a storage system, performing step e;

e. checking a list of available servers which require booting from the storage system;

if a server is available, performing step f;

f. matching the hardware attributes associated with the available server with those associated with the server image;

if the hardware attributes associated with the server image match with those associated with the available server, performing g;

g. assigning the server image to the available server; and

else performing step a;

else performing step a;

else performing step h;

h. booting the server with a server image on a local disk, the local disk being present on the server;

else performing step a;

else performing step a.

20. The method of claim 15, wherein booting servers includes the steps of:

a. broadcasting a request for a server image, the request being broadcasted by a server to the controller, the request being made using all installed Network Interface Cards (NICs), the NICs being installed on the server broadcasting the request;

b. assigning a default pre-boot image, the default pre-boot image being assigned by the controller to the server broadcasting the request;

c. downloading the pre-boot image, the pre-boot image being downloaded by the server;

d. running the pre-boot image;

e. sending a unique server identifier of the server and Medium Access Control (MAC) addresses of all the installed NICs, the unique server identifier of the server and MAC addresses being sent by the server to the controller;

f. checking for a server image corresponding to the server, the checking being

performed by the controller;

if a server image is assigned to the server and;

if the storage system is a local disk, performing step g;

g. booting the server with the assigned server image, the assigned server image being present on the local disk of the server;

else,

if the storage system is a Storage Area Network (SAN), performing steps h to j;

h. sending a virtual World Wide Name (WWN) to the server, the virtual WWN being sent by the controller to the server, the virtual WWN being used by the server to establish a connection with the SAN;

i. establishing a connection with the server image on the SAN, the connection being established by the server, the connection allowing access to the server image;

j. running the server image;

else, performing steps k to n;

k. running a pre-execution environment, the pre-execution environment being run by the server, the pre-execution environment residing on the server;

l. establishing a communication path between the server and a Network Attached Storage (NAS) device, the communication path being established using the virtual MAC address;

m. downloading the server image from the NAS device, the downloading being done by the server; and

n. running the downloaded server image;

else, performing step a.

21. The method of claim 16, wherein tracking the movement of servers across the ports of the layer 2 interconnect fabric includes the steps of:

a. waiting for a link-up event to be reported at the ports of the layer 2 interconnect fabric;

if a link-up event is reported at a port, performing step b;

b. getting a MAC address table from the switch for the port, the port reporting the link-up event;

if a MAC is found in the MAC address table, performing step c;

c. matching the MAC with a server configuration, the server configuration being stored in a configuration database in the controller database;

if a server configuration matches with the MAC in the MAC address table, performing step d;

d. checking if a server is in the port reporting a link-up event;

if a server is in the port reporting a link-up event and,

if the server in the port reporting a link-up event is same as the server with the matching configuration, performing step e;

e. retaining the server in the port reporting the link-up event;

else, performing steps f and g;

f. removing the server from the port;

g. moving the server into the port reporting the link-up event, the server having the configuration matching with the MAC, the MAC being found in the MAC address table; and

else, performing step g;

else, performing step h;

h. waiting for server to be discovered, the server being discovered by sending an information packet, the information packet being sent by the server, the information packet further comprising a MAC address, the MAC address corresponding to the NIC of the server being discovered;

else, performing step h;

else, performing step a.

23. The method of claim 17, wherein providing a substitute server includes the steps of:

a. waiting for a server failure to be reported;

if a server failure is reported and;

if a server removal is reported, the server being removed from a port and;

if the removed server has a server image assigned, performing steps b and c;

b. unassigning the server image from the server, the server reporting server failure; and

c. marking the server image as an available server image, the available server image requiring allocation;

else, performing step a;

else,

if the power of the server reporting server failure has been reported as off and;

if the server has a server image assigned, performing steps b and c;

else, performing step a;

else,

if the port being connected to the server reporting server failure has been reported as down, performing steps b and c;

else performing step a;

else performing step a.

24. The method of claim 15, wherein the method is implemented as a computer program product.

22. The method of claim 22, wherein discovering the server by sending an information packet includes the steps of:

a. waiting for an information packet to be received, the information packet being sent by a server;

if an information packet has been received, performing steps b and c;

b. receiving a unique server identifier and MAC addresses of the server, the unique server identifier and the MAC addresses of the server being present in the information packet;

c. matching the MAC addresses of the server with the MACs in the MAC address table of a switch, wherein the switch includes the port providing connectivity to the server;

if a MAC address out of the plurality of the MAC addresses of the server matches with the MACs received from the switch, performing step d;

d. moving the server into the port on which a matching MAC address is found;

if the server does not have an assigned server image and;

if the information packet indicates booting from a NAS device, performing step e;

e. marking the server as requiring a server image, the server image being stored on a NAS device;

else performing step f;

f. booting the server from a local disk, the local disk being present on the server.

else, performing step g;

g. running the assigned server image on the server.

else performing step a;

else, performing step a.