Abstract: Certain example embodiments relate to a computer system for performing a map reduce sequence. Nodes therein include at least one processor and memory and are divided into at least mapper and reducer nodes. Each mapper node executes a map function on input to generate intermediate output elements. Each said intermediate output element includes a first key-value pair. Each element key includes associated map and reduce task identifiers. Each element value includes substantive data, organized as another key-value pair. The intermediate output elements are stored to memory. Each reducer node: retrieves at least intermediate output element values from the memory of a given mapper node, using specified map and reduce task identifiers; stores the retrieved element values to its memory; executes a reduce function on the retrieved element values, in order; and outputs a result from the reduce function for the map reduce sequence. Disk operations advantageously are reduced or eliminated.

Abstract: An in-memory data grid (IMDG) server includes an i/o interface for transmitting and receiving data over a network. A random access memory (RAM) is configured with the IMDG which stores plural different data sets that may be requested for retrieval by applications that can connect to the IMDG server through the network. A processor receives requests for retrieval of data from the IMDG data sets, and sends the requested data to the requesting application. The IMDG server also includes a request prioritizer that determines, when two or more applications, which are competing applications that defined as having requests co-pending at the IMDG server to retrieve one or more of the data sets, which of the competing applications has the highest priority relative to the other competing applications. The request prioritizer causes a data set requested by the application with the highest priority to be handled before requests from the applications.

Abstract: Certain example embodiments relate to memory management techniques that enable users to “pin” elements to particular storage tiers (e.g., RAM, SSD, HDD, tape, or the like). Once pinned, elements are not moved from tier-to-tier during application execution. A memory manager, working with at least one processor, receives requests to store and retrieve data during application execution. Each request is handled using a non-transitory computer readable storage medium (rather than a transitory computer readable storage medium), if the associated data is part of a data cache that is pinned to the non-transitory computer readable storage medium, or if the associated data itself is pinned to the non-transitory computer readable storage medium. If neither condition applies, the memory manager determines which one of the non-transitory and the transitory computer readable storage mediums should be used in handling the respective received request, and handles the request accordingly.

Abstract: An in-memory data management (IMDM) system is described that includes an IMDM cluster, a load balancer, and a reallocation processor controller. The IMDM cluster includes a plurality of nodes. The controller determines whether there is an actionable load imbalance of existing data elements, stored among the plurality of nodes, based on a predefined criteria or rule. The controller further identifies a source node from which at least one data element is to be deleted and a target node to which the at least one data element that is to be deleted is to be added. The source node and target node contribute to the actionable load imbalance. The controller copies the at least one data element that is to be deleted from the source node into the target node, and deletes the at least one data element that is to be deleted from the source node.

Abstract: An in-memory data management (IMDM) system is described that includes an IMDM cluster, a load balancer, and a reallocation processor controller. The IMDM cluster includes a plurality of nodes. The controller determines whether there is an actionable load imbalance of existing data elements, stored among the plurality of nodes, based on a predefined criteria or rule. The controller further identifies a source node from which at least one data element is to be deleted and a target node to which the at least one data element that is to be deleted is to be added. The source node and target node contribute to the actionable load imbalance. The controller copies the at least one data element that is to be deleted from the source node into the target node, and deletes the at least one data element that is to be deleted from the source node.

Abstract: An in-memory data grid (IMDG) server includes an i/o interface for transmitting and receiving data over a network. A random access memory (RAM) is configured with the IMDG which stores plural different data sets that may be requested for retrieval by applications that can connect to the IMDG server through the network. A processor receives requests for retrieval of data from the IMDG data sets, and sends the requested data to the requesting application. The IMDG server also includes a request prioritizer that determines, when two or more applications, which are competing applications that defined as having requests co-pending at the IMDG server to retrieve one or more of the data sets, which of the competing applications has the highest priority relative to the other competing applications. The request prioritizer causes a data set requested by the application with the highest priority to be handled before requests from the applications.

Abstract: Certain example embodiments relate to a computer system for performing a map reduce sequence. Nodes therein include at least one processor and memory and are divided into at least mapper and reducer nodes. Each mapper node executes a map function on input to generate intermediate output elements. Each said intermediate output element includes a first key-value pair. Each element key includes associated map and reduce task identifiers. Each element value includes substantive data, organized as another key-value pair. The intermediate output elements are stored to memory. Each reducer node: retrieves at least intermediate output element values from the memory of a given mapper node, using specified map and reduce task identifiers; stores the retrieved element values to its memory; executes a reduce function on the retrieved element values, in order; and outputs a result from the reduce function for the map reduce sequence. Disk operations advantageously are reduced or eliminated.

Abstract: Certain example embodiments relate to memory management techniques that enable users to “pin” elements to particular storage tiers (e.g., RAM, SSD, HDD, tape, or the like). Once pinned, elements are not moved from tier-to-tier during application execution. A memory manager, working with at least one processor, receives requests to store and retrieve data during application execution. Each request is handled using a non-transitory computer readable storage medium (rather than a transitory computer readable storage medium), if the associated data is part of a data cache that is pinned to the non-transitory computer readable storage medium, or if the associated data itself is pinned to the non-transitory computer readable storage medium. If neither condition applies, the memory manager determines which one of the non-transitory and the transitory computer readable storage mediums should be used in handling the respective received request, and handles the request accordingly.