International Business Machines Corporation

Benjamin W Herta

Vinod Muthusamy

Vatche Isahagian

Jayaram Kallapalayam Radhakrishnan

Scott Boag

Atin Sood

Evelyn Duesterwald

Darrell Christopher Reimer

Avraham Ever Shinnar

Vijay Anand Saraswat

Debashish Saha

Punleuk Oum

Archit Verma

Gaodan Fang

David Cunningham

Chandan Mallik

J Kevin Lowe, Jr

A method of using a computing device to manage a lifecycle of machine learning models includes receiving, by a computing device, multiple pre-defined machine learning lifecycle tasks. The computing device manages executing a management-layer software layer for the multiple pre-defined machine learning lifecycle tasks. The computing device further generates and updates a machine learning pipeline using the management-layer software layer.

Configurable and scalable machine learning model lifecycle operator

A lifecycle management method, system, and computer program product include coordinating hardware, platform and application-level health checks for framework-independent and application-specific monitoring, failure detection, and recovery, coordinating the hardware, the platform, and the application-level health check by state-specific aggregation of distributed atomic status events, and creating a recovery policy based on the state-specific aggregation of the distributed atomic status events.

Scalable multi-framework multi-tenant lifecycle management of deep learning applications

A lifecycle management method, system, and computer program product include establishing a public key infrastructure (PKI) for end-to-end encryption of control plane and data plane communications by providing encryption between arbitrary components for applicant execution where an interaction pattern is isolated, secure, and a multi-tenant environment.

Executing a map reduce sequence may comprise executing all jobs in the sequence by a collection of a plurality of processes with each process running one or more mappers, combiners, partitioners and reducers for each job, and transparently sharing heap state between the jobs to improve metrics associated with the job. Processes may communicate among themselves to coordinate completion of map, shuffle and reduce phases, and completion of said all jobs in the sequence.

Transparent efficiency for in-memory execution of map reduce job sequences

The present invention provides a method, system and program product for invoking a remote method. Specifically, the invention provides an interface method object that may be used to map an application interface method call to a remote method call. The interface method object may be generated by retrieving an interface description for the remote method and using the interface description, including any necessary parameters determined from the interface description, to generate the interface method call. The interface method object may be generated dynamically at system runtime. The interface method object facilitates invocation of a remote method by mapping an interface method call from an application in interface method object format to a remote method call in the format of the remote method. Because the interface method call is in an interface method object format, changes to the underlying remote method do not affect the application's interface method call.

System, method and program product for invoking a remote method

In forming a cluster of processors and applications among a plurality of processors connected in a network the embodiment of a pair of cluster nodes, as applications, in each server/system and arranging for communication between them in a ring or tiered ring configuration of servers/systems provides network status monitoring and failure recovery in a highly redundant and flexible manner at increased speed without the requirement of separate communication links for monitoring and control or redundant hardware such as a so-called 'hot standby' processor in each server/system.

Software application cluster layout pattern

A method, system, and computer program product for continuous availability of non-persistence messages in a distributed platform are provided. The method includes providing a first client application independently in operative communication with at least two servers via a first client abstraction layer, generating a critical message by the first client application, duplicating the critical message at the first client abstraction layer for each of the servers, and transmitting the duplicate critical messages independently to each of the servers. The method further includes providing a second client application independently in operative communication with each of the servers via a second client abstraction layer, receiving one or more of the duplicate critical messages at the second client abstraction layer, selecting a critical message from the one or more duplicate critical messages, and delivering the selected critical message to the second client application.

Growing community of inventors

Poughquag, NY, United States of America

Benjamin W Herta