International Business Machines Corporation

Kyndryl, Inc.

Michael Man Behrendt

Gerd Breiter

Ruchi Mahindru

Mahesh Viswanathan

Harigovind Venkatraj Ramasamy

Soumitra Sarkar

Norbert George Vogl

Rafah A Hosn

Thomas Spatzier

Craig M Trim

Aaron K Baughman

Shikhar Kwatra

Utz Bacher

Dinakaran B Joseph

Lap Thiet Huynh

Linwood Hugh Overby, Jr

Andreas Bieswanger

Wesley McMillan Devine

Sivaram Gottimukkala

Georg Ochs

Stefan Pappe

Andrea Schmidt

Benjamin Blau

Steffen Haak

Rudi Studer

Clemens Van Dinther

Christof Weinhardt

Timo Kussmaul

Andreas Nauerz

Tamar Eilam

Einar Lueck

John Arthur Pershing, Jr

Gerhard Banzhaf

Michael H Kalantar

Christine Axnix

Oliver Koeth

Alexander V Konstantinou

Torsten Steinbach

Ismael Faro Sertage

Simon D Moser

Helmut H Weber

Dietmar Kuebler

Friedemann Baitinger

Michael S Law

Martin Reitz

Steffen Rost

Patrick Sullivan

Joachim Hagmeier

Jochen E Breh

Ruediger Schulze

Marc Schwind

Monika Illgner-Kurz

Kapil Jain

Harald Daur

Ruediger Maass

Natalie Speiser

Martin Henke

Stefan Wirag

Michael Haide

Gregor Möhler

Sachin Lingadahalli Vittal

Alexander Eckert

Manuela Kohler

Daniel Federschmidt

Dominik Froehlich

Johanna Ang'ani

Hans-Deiter Wehle

Claus Allwell

Sebastian Ott

Hans-Joachim Picht

Jens Wollenhaupt

Jr Overby

Wesley Mcmillan Devine

A system and related method reduce public cloud provisioning latencies using one or more processors. The method comprises, prior to receiving a volume provisioning request from a user, creating a pool of pre-provisioned generic volumes. The method further comprises receiving the request from the user to provision a volume from the pool, and then determining that a pre-provisioned generic volume is available for customization based on the request. Responsive to the determination, the method executes the actions comprising customizing the pre-provisioned generic volume based on the request, creating a custom volume, and providing the customized pre-provisioned volume to the user.

Pre-provisioning storage volumes in the public cloud to reduce provisioning latencies

One or more computer processors receive an address resolution protocol (ARP) request in response to a client connecting to a compute resource. The one or more computer processors identify an internet protocol (IP) address associated with the ARP request and the compute resource. The one or more computer processors determine the compute resource is unavailable or not ready for a connection with the client utilizing the identified IP address. The one or more computer processors, responsive to determining that the compute resource is unavailable, provision and deploy a new compute resource, wherein the new compute resource is available to the client under the identified IP address.

Address resolution protocol based deployment

Aspects of the present invention disclose a method, computer program product, and system for query execution in a multi-tenant cloud service. The method includes one or more processors determining category classes for service queries. The method further includes sending for execution, a selected number of service queries from one of the determined category classes to a shadow query engine. Respective service queries of the categorically classified service queries comprise a different set of configuration parameter values for the shadow query engine. The method further includes recording metadata for the selected number of service queries of the one category class executed on said shadow query engine. The method further includes determining correlations between the recorded metadata. The method further includes determining, from the determined correlations, optimal configuration parameter values comprising optimal configuration parameters for executing the selected number of service queries of the one category classes.

Shadow experiments for serverless multi-tenant cloud services

One or more computer processors receive a domain name system (DNS) request in response to a client connecting to a compute resource. The one or more computer processors decoding the DNS request into one or more provision parameters. The one or more computer processors determining that the compute resource is unavailable for a connection with the client utilizing the identified IP address. The one or more computer processors, responsive to determining that the compute resource is not available or not ready, provisioning and deploying a new compute resource according to the one or more decoded provision parameters, wherein the new compute resource is available to the client under the identified IP address.

Domain name based deployment

An autonomic function executing in an artificial intelligence environment determines that a fused model responsive to a new problem space has below a threshold level of accuracy in the new problem space. A spliced layer in the fused model is autonomically cloned, the spliced layer having been extracted from a second model and inserted at a location in the fused model. The cloned layer is autonomically inserted at a second location in the fused model. An automatically constructed vector transformation transforms an output vector of a previous layer in an immediately previous location in the model relative to the second location. The cloned layer is automatically fused in the fused model using the transformed output vector as input to the cloned layer, forming a deep fused model that has a revised accuracy that is higher than the accuracy relative to an ontology of the new problem space.

Autonomic vertical deepening in neural networks transfer learning

An autonomic function is caused to execute in an artificial intelligence environment to detect a new problem space. Using the autonomic function, a first model is selected. The first model includes a first trained neural network corresponding to a first ontology. A second model is automatically identified. the second model includes a second trained neural network corresponding to a second ontology. A layer is autonomically extracted from the second model and inserted into a location in the first model. A vector transformation is automatically constructed to transform an output vector of a previous layer in an immediately previous location in the model relative to the location. The layer is automatically fused in the first model using the transformed output vector as input to the layer, the fusing forming a fused model that is operable on an ontology of the new problem space.

Autonomic horizontal exploration in neural networks transfer learning

Embodiments of the present invention provide systems and methods for constructing a plan for creating a cloud service. In one embodiment, a configurator receives a request for one or more services making up a cloud service, and a preliminary plan for the requested services is generated. A service provider determines that it is able to fulfill a requested service, and inserts a sub plan for fulfilling the requested service into the preliminary plan. A final plan is generated, which includes a sub plan from each service provider inserted into the preliminary plan.

Dynamic construction of cloud services

Deploying a software service. A defined software service offering is selected from a plurality of defined software service offerings. An instantiation order for deploying the selected defined software service offering on a corresponding computer system is generated. Each defined software service offering comprises a software service definition and at least one defined software policy. One selected from the group consisting of the software service definition, the software policy, and a combination thereof, provides a hardware configuration for the software service, a software configuration for the software service, and a managing policy for the software service. The instantiation order of the software service for execution on the computer system is created based on the hardware configuration, the software configuration and the managing policy.

Method, arrangement, computer program product and data processing program for deploying a software service

An order processing system can include an order processing container, a factory registry, a relationship registry, and synchronization function component. The order processing system can handle orders, which are build plans including a set of tasks. The tasks can specify programmatic actions which may include creation, deletion, and modification of resources and resource topologies. The order processing container can be central engine that programmatically drives order processing actions. The factory registry can support a creation and deletion of resource instances in a resource topology defined by at least one order. The relationship registry can maintain relationships among resources. The synchronization function component can permit transparent usage of shared resources in accordance with shared usage resource topology parameters specified within processed orders.

Synchronizing shared resources in an order processing environment using a synchronization component

Event-driven processing is facilitated by a unikernel-based method which includes receiving, based on an event, a code execution request containing parameter values for an application code unikernel. The application code unikernel includes uploaded application code in a single-purpose virtual machine image. Based on receiving the code execution request, the application code unikernel is run, which includes providing the parameter values as bootvariables as part of booting the program code unikernel. Further, running the application code unikernel may include executing the application code unikernel directly on a hypervisor. The application code unikernel may include only operating system components required to run the uploaded application code.

Facilitating event-driven processing using unikernels

A system, and computer program product for managing cloud services are provided in the illustrative embodiments. A first set of management functions is instantiated external to a microkernel of a management system for managing a set of cloud services. The management system further includes a second set of management functions executing within the kernel of the management system. A subset of the first set of management functions is associated with a cloud service in the set of cloud services. A management task for the cloud service is handled using the subset of the first set of management functions. A usage information corresponding to the use of the subset of the first set of management functions for handling the management task for the cloud service is reported.

Managing cloud services

A first set of management functions is instantiated external to a microkernel of a management system for managing a set of cloud services. The management system further includes a second set of management functions executing within the kernel of the management system. A subset of the first set of management functions is associated with a cloud service in the set of cloud services. A management task for the cloud service is handled using the subset of the first set of management functions. A usage information corresponding to the use of the subset of the first set of management functions for handling the management task for the cloud service is reported.

A mechanism is provided for handling incidents occurring in a managed environment. An incident is detected in a resource in the managed environment. A set of incident handling actions are identified based on incident handling rules for an incident type of the incident. From the set of incident handling actions, one incident handling action is identified to be executed based on a set of impact indicators associated with the set of incident handling rules. The identified incident handling action is then executed to address the failure of the resource.

Automated fault and recovery system

A method, an apparatus and an article of manufacture for detecting an event storm in a networked environment. The method includes receiving a plurality of events via a plurality of probes in a networked environment, each of the plurality of probes monitoring a monitored information technology (IT) element, aggregating the plurality of events received into an event set, and correlating the plurality of events in the event set to determine whether the plurality of events are part of an event storm by determining if the plurality of events in the event set meet one or more event storm criteria.

Real-time event storm detection in a cloud environment

Growing community of inventors

Randersacker, Germany

Michael Man Behrendt