International Business Machines Corporation

Peter Thomas Donovan

Simon C Chu

Richard Alan Dayan

William W Buchanan, Jr

Gregory Brian Pruett

David Michael Green

Paul Brian Tippett

William Paul Zeggert

Karl Ross Shultz

John Michael Brantly

Akram Abboud

Peter Gerard Chin

Timothy J Green

Thomas William Lange

Eric Richard Kern

Charles Robert Lefurgy

Malcolm Scott Allen-Ware

David Benson Rhoades

Guillermo Jesus Silva

Joseph Edward Bolan

Benjamin Russell Grimes

Andrew Hamilton Wray

Frank Cassatt Harwood

Jeffrey Wayne Anderson

Robert K Christner

Paul Brian Tippet

David Ohlemacher

Balaji Ramamoorthy

Controlling server power usage in a data center is provided. Power usage among a plurality of server racks in active mode processing a set of workloads in the data center is managed. It is detected that a new server rack in standby mode is being added to the plurality of server racks. It is ensured that the new server rack in the standby mode is properly controlled and monitored prior to transitioning the new server rack to the active mode. It is determined whether power safety criteria are met to safely join the new server rack to the plurality of server racks prior to transitioning the new server rack from the standby mode to the active mode. The new server rack is transitioned to the active mode in without exceeding a power budget for the plurality of server racks in response to determining that the power safety criteria are met.

Elastic resource control in a container orchestration environment

Identifying an operating system running on a computer system. In one aspect of the invention, an enumeration pattern is collected, the enumeration pattern describing an enumeration of a device that has been performed between the device and the operating system running on a host computer system. The type of the operating system running on the host computer system is identified based on the collected enumeration pattern.

Method for using device enumeration information to identify an operating system running on a computer system

A system remotely boots a plurality of server blades in a blade center to update each blade. A temporary network is created to provide an out-of-band communication channel between a management module and the plurality of server blades. The management module receives and saves the original boot sequence for each server blade and alters each boot sequence such that each server blade begins with a network boot. After receiving an initial transmission over the network containing server blade boot instructions and storing the boot instructions in a temporary boot server, subsequent server blade boot requests from the network are intercepted and the stored boot instructions are re-routed from the temporary boot server to each of the remaining blades in the blade server thus removing the need for repeated boot packet transmission over the network.

Server blade network boot method that minimizes required network bandwidth

Aspects for detection and correction of entrance into an improper master boot record (MBR) state in a computer system are described. The aspects include providing a proxy MBR at a predetermined location on a storage device of a computer system to provide a proxy MBR. The proxy MBR is then utilized during a restart operation. An original MBR maintained at an original location on the storage device is then utilized during a reboot operation when the utilization of the proxy MBR is unsuccessful.

Method and system for detection and correction of entrance into an improper MBR state in a computer system

Disclosed is a method for automatically re-provisioning an appliance server without significant user-interaction. The disk drive of a server to be utilized as an appliance server is partitioned into at least three partitions, including a systems partition, a network operating system partition, and an images partition. A re-provisioning utility is provided, which, when activated by a user, removes the present application from the server and forwards it to a storage location on the network or in a fourth storage partition and installs another application on the server. The system is re-booted and re-configured to provide full operation of the new application.

Automatic appliance server re-provision/re-purposing method

Disclosed is a method and hard disk configuration for accommodating different sizes of applications during an automatic re-provisioning of an appliance server. The disk drive of the appliance server is partitioned with a system partition, a network operating system (NOS) partition, a float partition, and an images partition. The float partition is utilized to provide additional space to the NOS partition and the images partition, when required. A re-provisioning utility is provided, which initiates both a create image utility and an apply image utility, whereby an image file of a current application and associated operating system is created and a stored image file of a second application is installed on the appliance server. When the apply image utility is initiated, the NOS partition is dynamically extended into the float partition server if the second application requires more space than is provided in the NOS partition. Similarly, when an image file is stored in the images partition and the image file requires more space than is available in the images partition, the images partition is dynamically extended into the float partition.

Appliance server with a drive partitioning scheme that accommodates application growth in size

Growing community of inventors

Cary, NC, United States of America

Peter Thomas Donovan