International Business Machines Corporation

Frequentz LLC

Siemens Aktiengesellschaft

Shrinivas Kulkarni

Rachit Arora

Dharmesh Jain

Anjali Agarwal

Padmanabhan Krishnan

Rakhi S Arora

Deepashree Gandhi

Subin Shekhar

Karin Murthy

Naveenkumar V Muguda

Gary J Robinson

Nisanth M Simon

Siba Prasad Satapathy

Christian C Clauss

Anthony C Asher

Steffen Schaefer

Raha Debarshi

Bharat K Gera

Sunil R Ganatra

Samrat Bhattacharjee

Disclosed embodiments provide cluster creation using self-joining cluster nodes. Each node within a cluster is assigned a role of master or non-master. A master node initializes and creates a cluster. A non-master, or 'role' node initializes and joins a cluster established by the master. If the master has not yet initialized, the non-master node waits until the master node is operational and then joins the cluster. A description file associated with each node describes the services and configuration required for the node's assigned role. In this way, non-master nodes can be spawned independently of the master node, providing a greater level of scalability, improved fault tolerance, and increased availability.

Cluster creation using self-aware, self-joining cluster nodes

The disclosed embodiments include a system, computer program product, and computer-implemented method configured to restore binaries of a container. In one embodiment, the computer-implemented method receives a container install request to install a container on a host system. The computer-implemented method determines binaries that were previously installed with the container. The computer-implemented method installs the container and the binaries that were previously installed with the container in response to the container install request.

Binary restoration in a container orchestration system

A mechanism is provided in a data processing system for creating clusters on demand. The mechanism installs a cluster on the data processing system. The cluster comprises a master node and a managed node having distributed software installed thereon. The mechanism stores state of cluster on external volumes and removes references specific to the cluster from the state of the cluster stored on the external volumes. The mechanism takes a snapshot of the state of the cluster stored on the external volumes to form a set of volume templates and takes a snapshot of images of the master node and the managed node to form a set of node type images. Responsive to receiving a request to create a specified cluster, the mechanism creates a new set of nodes on the data processing system based on the set of node type images, clones the set of volume templates to form a new set of volumes in the data processing system, mounts the new set of volumes to the new set of nodes, and recreates configurations based on details in the request.

Optimized creation of distributed storage and distributed processing clusters on demand

A mechanism is provided for automatically recovering one or more distributed cluster nodes on a host system. Responsive to the host system recovering, restoring, or restarting, a set of distributed clustered systems affected due to the reboot of the host system and details associated with each of the set of distributed clustered systems are identified. Using the details, a set of nodes that operate on the host system are identified. For each node: respective components and configurations are prepared; a heartbeat is initiated thereby causing the node to register with a cluster network and identify peers within their associated distributed clustered system; a determination is made as to whether the node has reestablished connection with its associated distributed clustered system; and, responsive to the node reestablishing connection with its associated distributed clustered system, service components are started on the node.

Self-recoverable multitenant distributed clustered systems

A mechanism is provided for allocating a hyper-threaded processor to nodes of multi-tenant distributed software systems. Responsive to receiving a request to provision a node of the multi-tenant distributed software system on the host data processing system, a cluster to which the node belongs is identified. Responsive to the node being a second type of node, responsive to determining that another second type of node in the same cluster has been provisioned on the host data processing system, and responsive to the number of unallocated VPs on different physical processors from that of the other second type of node being greater than or equal to the requested number of VPs for the second type of node, the requested number of VPs for the second type of node is allocated each to a different physical processor from that of the other second type of node.

Hyper-threaded processor allocation to nodes in multi-tenant distributed software systems

Map reduce processing is performed holistically, to yield map reduce results data, on an aggregated group of data sets, which had been generated and maintained with 'splits' separating the separate date sets, rather than being performed separately on the split data sets and then aggregating (or attempting to aggregate) the map reduce processing results.

Smarter big data processing using collaborative map reduce frameworks

A plurality of subgroups of identifiers are identified including at least a first subgroup of identifiers and a second subgroup of identifiers within an identifier range. A first set of identifiers is selected in the first subgroup. The first set of identifiers in the first subgroup is compared with a second set of identifiers in the second subgroup and a first set of common identifiers is identified using metadata of the second set of identifiers. A third set of identifiers is generated by eliminating the first set of common identifiers from the first set of identifiers. The third set of identifiers is compared with a fourth set of identifiers in a repository of stored identifiers, to identify a second set of common identifiers. A fifth set of identifiers is generated by eliminating the second set of common identifiers from the third set of identifiers.

Allocating unique identifiers using metadata

Techniques for unifying data are provided. The techniques include transforming a query into one or more sub-queries that can be answered by one or more types of data services, and wherein the query touches one or more joins across data residing in one or more enterprise systems, querying one or more target data services for each of the one or more sub-queries, aggregating one or more sets of results based on the one or more target data services, and unifying the one or more sets of aggregated results.

Unifying hetrogenous data

Methods and apparatus, including computer program products, implementing and using techniques for providing a discovery service that provides a mapping from an electronic product code to electronic product code information services that include information about an item represented by an electronic product code. The discovery service includes a database containing a list of custodians for the item and a service interface through which electronic product code information services and applications can access the entries in the list. The list includes an entry for each custodian. Each entry includes the electronic product code number, a pointer to an electronic product code information service associated with the custodian, and a trip identifier for the item. The service interface includes several services enabling traceability of the item represented by the electronic product code. Methods and apparatus, including computer program products for using a trip identifier in a discovery service are also described.

Enhancing discovery services in an EPC network

A system and method for role pen based messaging in a collaborative environment is presented. Online users utilize clients and a server to collaborate in a distributed presentation. One of the users is a presentation leader and sends presentation content to the server which distributes the presentation content to participating users. One of the users wishes to send a comment to a subset of the participating users. The user selects a role pen from a role pen stand whereby the role pen corresponds to a subset of the participating users. The user then generates a comment using his keyboard and/or mouse pointer, and the user's client sends the comment to recipient users' clients that correspond to the role pen. The recipient users' clients display the comment in the same window as the distributed presentation. Participating users that do not correspond to the role pen do not receive the comment.

Growing community of inventors

Bengaluru, India

Shrinivas Kulkarni