Computer cluster and method for providing a disaster recovery functionality for a computer cluster

• US 8,843,624 B2
• Filed: 12/10/2010
• Issued: 09/23/2014
• Est. Priority Date: 12/11/2009
• Status: Active Grant

• Current Assignee: Deutsche Telekom AG
• Original Assignee: Deutsche Telekom AG
• Inventors: Britsch, Matthias, Fuchs, Georg, Hirsch, Colin, Florath, Andreas
• Primary Examiner(s): Chan, Wing F
• Assistant Examiner(s): Thieu, Benjamin M
• Application Number:
  US13/515,035
• Publication Number:
  US 20130013766A1
• Time in Patent Office: 1,383 Days
• Field of Search: 709/1, 709/224, 709/221, 709/223, 709/201
• US Class Current: 709/224
• CPC Class Codes: G06F 11/203 using migration
• CPC Class Codes: G06F 11/2038 with a single idle spare processing component
• CPC Class Codes: G06F 2009/45575 Starting, stopping, suspending or resuming virtual machine instances
• CPC Class Codes: G06F 2009/45591 Monitoring or debugging support
• CPC Class Codes: G06F 2009/45595 Network integration; Enabling network access in virtual machine instances
• CPC Class Codes: G06F 9/4401 Bootstrapping security arrangements therefor G06F21/57
• CPC Class Codes: G06F 9/45558 Hypervisor-specific management and integration aspects
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A computer cluster includes: a first compartment, further including a plurality of first physical nodes configured to provide first resources for guest virtual instances, each of the plurality of first physical nodes further including a mass storage device, the mass storage device further including: a first partition provided for booting the first physical node; and a second partition as part of the first resources; and a second compartment, further including a plurality of second physical nodes configured to provide second resources for guest virtual instances, each of the plurality of second physical nodes further including a mass storage device, the mass storage device including: a third partition provided for booting the second physical node; and a fourth partition as part of the second resources. The first and the second physical nodes are further configured to provide third resources for a plurality of storage management processes.

Claims

1. A distributed architecture for hosting a server-based application in a virtualized environment inside a cluster of hardware hosts, wherein the same sever-based application is hosted on different hardware hosts of the cluster of hardware hosts, wherein the distributed architecture comprises:

virtual instances, configured to run on a hardware host of the cluster of hardware hosts, the virtual instances being connected to external storage, wherein the virtual instances are further configured to be restarted on another hardware host of the cluster of hardware hosts in the event that the hardware host fails;

a controller comprising a processor, the controller being configured to connect to storage resources, which are enabled for dynamic creation and resizing during operation without interfering with availability or uptime of a virtual instance to which the storage resources are assigned, and which are accessible via network protocols so as to cope with loss of paths or storage redundancy in a multipath configuration of an internet Small Computer System Interface (iSCI) in which an iSCI initiator writes over more than one path to a physical storage resources;

first agents, wherein the first agents are application programs within the virtual instances, the first agents being configured to receive instructions from the controller, to act on directives issued by the controller, and to report a status of the virtual instances to the controller periodically;

second agents, wherein the second agents are of a standalone node of a computer cluster, the second agents being configured to receive instructions from the controller and to act on directives issued by the controller;

ports of the standalone node, which are reserved for providing connectivity to external appliances;

control domains with which the hardware hosts of the cluster of hardware hosts are equipped, the control domains being configured to require an infrastructural process embedded into the control domain, wherein the infrastructural process acts;

in case of a disaster, or

during a reboot of the local control domain or parts of the hardware hosts of the cluster of hardware hosts, or

in the wake of hardware faults,

or any combination thereof;

a communication system comprising a processor, the communication system being included in the control domains, the communication system being configured to contact other control domains and to distribute configuration data over all the control domains, with the configuration data being stored in databases of flat file systems in the control domains;

a management system comprising a processor, wherein the management system is logically superior to the control domains, the management system being enabled to act as a single redundancy controller by being configured to restart the virtual instances on the another hardware host of the cluster of hardware hosts that it has access to, the management system further being multi-tenancy enabled and able to run inside one of the virtual instances in the cluster of hardware hosts that it manages, and the management system further being configured to enable creation of elements in a data center centrally to maintain a configuration database;

a first compartment comprising a plurality of first physical nodes configured to provide first resources for guest virtual instances, each of the plurality of first physical nodes comprising a first mass storage device, the first mass storage device comprising;

a first partition provided for booting the first physical node; and

a second partition as part of the first resources; and

a second compartment comprising a plurality of second physical nodes configured to provide second resources for guest virtual instances, each of the plurality of second physical nodes comprising a second mass storage device, the second mass storage device comprising;

a third partition provided for booting the second physical node; and

a fourth partition as part of the second resources;

wherein the first and the second physical nodes are configured to provide third resources for a plurality of storage management processes;

wherein each of the plurality of first physical nodes and the plurality of second physical nodes is provided for at least one of the plurality of storage management processes; and

wherein the storage management processes are configured for monitoring each other.

2. The distributed architecture according to claim 1,

wherein the first compartment comprises a first militarized zone and a first demilitarized zone, and the second compartment comprises a second militarized zone and a second demilitarized zone;

wherein in the first militarized zone only a restricted data exchange compared to the first demilitarized zone is allowed;

wherein in the second militarized zone only a restricted data exchange compared to the second demilitarized zone is allowed;

wherein each first physical node in the first militarized zone comprises a fifth partition as part of fourth resources for a first administrative virtual instance; and

wherein each second physical node in the second militarized zone comprises a sixth partition as part of fifth resources for a second administrative virtual instance.

3. The distributed architecture according to claim 2, wherein the plurality of fifth partitions are connected in a RAID1 system and the plurality of sixth partitions are connected in a further RAID1 system.

4. The distributed architecture according to claim 1, wherein the first resources and the second resources are provided for at least one telecommunication application.

5. The distributed architecture according to claim 1, wherein the storage management processes provide for disaster recovery.

6. The distributed architecture according to claim 2, wherein

one of the first administrative virtual instance and the second administrative virtual instance is configured to act as a master virtual instance with the other of the first administrative virtual instance and the second administrative virtual instance being configured to act as a slave virtual instance when both the first administrative virtual instance and the second administrative virtual instance are operational;

and

at least one of the first administrative virtual instance and the second administrative virtual instance is configured to act as a master virtual instance with the other of the first administrative virtual instance and the second administrative virtual instances being configured to be activated and act as a slave virtual instance when only one of the first administrative virtual instance and the second administrative instance is operational.

7. A method for providing disaster recovery functionality for the distributed architecture according to claim 1, the method comprising:

operating at least one of a first administrative virtual instance and a second administrative virtual instance, wherein the first administrative virtual instance corresponds to a first compartment of the computer cluster and the second administrative virtual instance corresponds to a second compartment of the computer cluster; and

when both the first administrative virtual instance and the second administrative virtual instance are operational;

operating one of the first administrative virtual instance and the second administrative virtual instance as a master virtual instance; and

operating the other of the first administrative virtual instance and the second administrative virtual instance as a slave virtual instance;

when only one of the first administrative virtual instance and the second administrative instance is operational;

operating the operational administrative virtual instance as a master virtual instance;

activating the other administrative virtual instance; and

operating the other administrative virtual instance as a slave virtual instance.