Dell Products Inc.

Ananda Chinnaiah Sankaran

Nam Van Nguyen

Bharath Vasudevan

Sumankumar Singh

Kong Yang

A method to handle peak database workloads may include requesting resources, receiving virtual-machine information in response to the requesting, allocating first and second portions of a workload according to the virtual-machine information, processing the first portion on a virtual machine to generate a first result, creating a cloned virtual machine with a virtualization layer, and cloning with the virtualization layer a storage allocated to the virtual machine to create a cloned storage.

Method to dynamically provision additional computer resources to handle peak database workloads

A method to handle peak database workloads is disclosed. In one form of the disclosure, the method can include requesting resources, receiving virtual-machine information in response to requesting, and allocating first and second portions of a workload according to the virtual-machine information. The method can also include processing the first portion on a virtual machine to generate a first result, processing the second portion on a cloned virtual machine to generate a second result, and aggregating the first and the second results to form a response.

A server cluster comprises a physical node and a virtual host node. The physical node includes an active node running an application. The virtual host node includes an inactive virtual node. The virtual node is activated upon failure of the active node and the application is failed over to the virtual node.

On-demand provisioning of computer resources in physical/virtual cluster environments

A method and systems for recovering from a failure in a virtual machine are provided. In accordance with one embodiment of the present disclosure, a method for recovering from failures in a virtual machine is provided. The method may include, in a first physical host having a host operating system and a virtual machine running on the host operating system, monitoring one or more parameters associated with a program running on the virtual machine, each parameter having a predetermined acceptable range. The method may further include determining if the one or more parameters are within their respective predetermined acceptable ranges. In response to determining that the one or more parameters associated with the program running on the virtual machine are not within their respective predetermined acceptable ranges, a management module may cause the application running on the virtual machine to be restarted.

System and method of recovering from failures in a virtual machine

According to various illustrative embodiments of the present invention, a method for adaptive cluster input/output control includes starting a nonessential input/output operation using a first controller in a first node of a cluster, informing at least a second controller in a second node of the cluster about starting the nonessential input/output operation, and increasing the nonessential input/output operation by a predetermined utilization percentage. The method also includes waiting for a predetermined amount of time, determining whether the nonessential input/output operation has been completed, and determining whether the at least the second controller in the second node has substantially decreased performance. The method also includes decreasing the nonessential input/output operation by the predetermined utilization percentage if the nonessential input/output operation utilization percentage is greater than the predetermined utilization percentage and if the at least the second controller in the second node has substantially decreased performance, and informing the at least the second controller in the second node of the cluster about the completion of the nonessential input/output operation if the nonessential input/output operation has been completed.

Adaptive input/output bus sharing method to improve performance of SCSI/SAS clusters

A SAN-based cluster backup system and method are provided. The system and method are automated, do not use a LAN for backup data, and are made aware of application failover events. The system and method are composed of two main components: a backup service, and a primary coordinator. The backup service performs the backup of the applications that are hosted on a particular node. The backup service periodically checkpoints the state of the backup job and communicates the status to the primary coordinator. The primary coordinator controls all backup operations in the cluster. The user submits backup jobs for the applications through the primary coordinator. If a node fails during a backup operation, the primary coordinator can ensure that the failed backup job can be resumed from the last checkpoint on the failed-over node. In this way, repetitive backups can be avoided, thereby increasing efficiency.

Growing community of inventors

Austin, TX, United States of America

Ananda Chinnaiah Sankaran