Remote Management Of Distributed Datacenters

• US 20130318242A1
• Filed: 05/24/2012
• Published: 11/28/2013
• Est. Priority Date: 05/24/2012
• Status: Active Grant

• Current Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Original Assignee: ScaleXtreme, Inc. (Cloud Software Group)
• Inventors: Srinivasa, Balaji
• Granted Patent: US 10,235,205 B2
• Time in Patent Office: Days
• US Class Current: 709/226
• CPC Class Codes: G06F 9/5011 the resources being hardware resources other than CPUs, Servers and Terminals
• CPC Class Codes: H04L 2012/5631 Resource management and allocation
• CPC Class Codes: H04L 41/042 comprising distributed management centres cooperatively managing the network
• CPC Class Codes: H04L 61/2592 using tunnelling or encapsulation
• CPC Class Codes: H04L 65/1066 Session management
• CPC Class Codes: H04L 9/3268 using certificate validation, registration, distribution or revocation, e.g. certificate revocation list [CRL]

Abstract

A cloud-based management system for distributed datacenters provides remote management of the datacenters from terminals at various locations. A virtual cloud-based operating system facilitates a transparent system call architecture, permitting subscribers to deploy, configure and manage operations at machines operating on different local operating systems. The virtual operating system can canonicalize available resources into a namespace to expose the resources as files at terminals. Asynchronous execution of subscriber commands across a group of target machines is provided with a notification procedure to control bandwidth and resource allocation. A file distribution model without a central controller for a group of target machines is provided to alleviate interaction with the management cluster for file distribution.

Claims

1. A computer-implemented method of remote datacenter management, comprising:

receiving at a management server a resource request associated with a group of target machines at a datacenter;

assigning a transaction identifier to the resource request;

determining a subscriber identifier associated with the resource request and a machine identifier for each target machine;

receiving at a discovery server a resource request from an agent on each target machine; and

in reply to the resource request from each agent at the discovery server, providing a reply including the transaction identifier and an instruction to issue a resource request to the management server.

2. A computer-implemented method according to claim 1, further comprising:

receiving at the management server the resource request from each agent including the transaction identifier

3. A computer-implemented method according to claim 2, further comprising:

matching the resource request from each agent to the resource request from the subscriber based on the transaction identifier.

4. A computer-implemented method according to claim 3, further comprising:

providing a resource reply in response to the resource request from each agent including a resource from the resource request from the subscriber.

5. A computer-implemented method according to claim 4, wherein the resource is a system call.

6. A computer-implemented method according to claim 4, wherein the resource is an operation to be executed across the group of target machines.

7. A computer-implemented method according to claim 4, wherein the resource is a file to be distributed across the group of target machines.

8. A computer-implemented method according to claim 2, wherein the resource request from each agent includes the transaction identifier and an identifier of the target machine.

9. A computer-implemented method according to claim 1, wherein:

the management server includes a first management server at which the resource request from the subscriber is received;

the management server includes a second management server, the instruction in the reply to the resource request from each agent is to issue a resource request to the second management server.

10. A computer-implemented method according to claim 1, wherein:

the management server includes a first management server at which the resource request from the subscriber is received;

the management server includes a plurality of additional management servers, the instruction in the reply to the resource request from each agent is to issue a resource request to one of the plurality of additional management servers.

11. A computer-implemented method according to claim 1, wherein:

the resource request associated with the group of target machines includes a communication protocol packet; and

the resource request from each agent includes a communication protocol packet.

12. A method according to claim 11, wherein:

the communication protocol packet of the resource request associated with the group of target machines is an internet protocol packet;

the communication protocol packet of the resource request from each agent is an internet protocol packet;

13. A method according to claim 12, wherein the internet protocol packet is an HTTP packet.

14. One or more processor readable storage media having processor readable code embodied on the one or more processor readable storage media, the processor readable code for programming one or more processors to perform a method comprising:

receiving a system call associated with an operating system of a client device;

determining whether the system call references remote resources;

if the system call does not reference remote resources, providing the system call to the operating system;

if the system call references a remote resource, converting the system call to a system protocol packet;

attaching the system protocol packet to a communication protocol packet; and

providing the communication protocol packet to a remote management cluster.

15. One or more processor readable storage media according to claim 14, wherein attaching the system protocol packet to the communication protocol packet comprises encapsulating the system protocol packet into a header of the communication protocol packet.

16. One or more processor readable storage media according to claim 14, wherein the method further comprises:

receiving at the client device, a second communication protocol packet from the remote management cluster, the second communication protocol packet including a response code for the system call corresponding to execution of the system call at a first target machine.

17. One or more processor readable storage media according to claim 16, wherein the method further comprises

accessing a second system protocol packet in the second communication protocol packet;

accessing the response code from the second system protocol packet; and

providing the response code to a terminal at the client device.

18. One or more processor readable storage media according to claim 17, wherein the method further comprises:

receiving by an agent at the first target machine a third communication protocol packet in response to a request issued to the remote management cluster;

determining the system call from the third communication protocol packet; and

providing a result of execution of the system call at the target machine in a fourth communication protocol packet.

19. One or more processor readable storage media according to claim 18, wherein the method further comprises

decoding a third system protocol packet from the third communication protocol packet;

determining the system call from the third system protocol packet;

converting the result into a fourth system protocol packet; and

attaching the fourth system protocol packet to the fourth communication protocol packet.

20. A system for remote management of datacenters using internet protocol communication, comprising:

at least one management server configured to;

generate a base certificate for a group of target machines at a remote datacenter, the base certificate including information for a subscriber of data management services;

send the base certificate to the group of target machines;

generate a unique certificate for each target machine in response to receiving the base certificate from an agent at each target machine; and

provide the unique certificate to each corresponding target machine; and

a discovery server configured to;

receive from the agent at each target machine a resource request; and

in response to an instruction from the at least one management server indicating a subscriber request for the group of target machines, generate a reply to the agent with a transaction identifier for the subscriber request.

21. A system according to claim 20, wherein the at least one management server includes:

a first management server that generates the base certificate and the unique certificate; and

a second management server configured to receive the subscriber request;

wherein the reply to the agent includes an address of the second management server.

22. A computer-implemented method for remote management of datacenters, comprising:

receiving from a subscriber a request for execution of an operation across a group of target machines;

generating a unique tag for the operation;

receiving a request from an agent at each target machine using a communication protocol;

providing the operation to each target machine;

generating a reply to a resource request from an agent of each target machine, the reply including a communication protocol packet having the unique tag and an instruction to execute the operation at the corresponding target machine.

23. A computer-implemented method according to claim 22, wherein the instruction includes a location of a tracker file for an operation file to be executed on each target machine.

24. A computer-implemented method according to claim 22, wherein the instruction includes the operation to be executed on each target machine.

25. A computer-implemented method according to claim 22, wherein the unique tag for the operation is different for each target machine.

26. A computer-implemented method according to claim 22, wherein the communication protocol packet is a first communication protocol packet, the method further comprising:

receiving a second communication protocol packet at the management server from the agent of each target machine, the second communication protocol packet including a status of execution of the operation at the target machine.

27. A computer-implemented method according to claim 26, wherein the second communication protocol packet is received as part of a resource request from the agent at each target machine.

29. A computer-implemented method according to claim 22, wherein the instruction includes a location of a tracker file for an operation file to be executed on the target machine.

30. A computer-implemented method according to claim 29, wherein the target machine is a first target machine, the method further comprising:

accessing the tracker file;

accessing a first portion of the operation file at a second target machine based on the tracker file; and

updating the tracker file to reflect that that the first target machine includes the first portion of the operation file.

31. A computer-implemented method according to claim 30, wherein:

accessing the tracker file comprises accessing the tracker file from a third target machine.

32. A computer-implemented method according to claim 30, wherein

accessing the tracker file comprises accessing the tracker file from the management server using one or more communication protocol packets.

28. A computer-implemented method for remote datacenter management, comprising:

receiving by an agent at a target machine a communication protocol packet from a management server, the communication protocol packet having a unique tag for a subscriber request submitted to the management cluster and an instruction to execute an operation at the target machine in accordance with the subscriber request;

creating a directory at the target machine based on the unique tag;

providing an output associated with execution of the operation at the target machine in the directory; and

generating a communication protocol packet for the management server having the unique tag and the output associated with the operation at the target machine.

33. A computer-implemented method according to claim 33, wherein:

the output associated with the operation is a status file associated with the operation.

34. A computer-implemented method according to claim 33, wherein the output is a first output, the method further comprising:

providing a second output associated with execution of the operation at the target machine in the directory, the second output including a debug file associated with execution of the operation at the target machine.