International Business Machines Corporation

Hao Yu

Joefon Jann

Pratap Chandra Pattnaik

Hubertus Franke

Jose Eduardo Moreira

Kattamuri Ekanadham

Ramanjaneya Sarma Burugula

Niteesh Kumar Dubey

Manoj N Kumar

Giora Biran

Mauricio J Serrano

Ilie G Tanase

William Pettit Horn

Amit Golander

Brian Mitchell Bass

Kun Wang

Asser Nasreldin Tantawi

Seetharami R Seelam

Laurent Schares

I-Hsin Chung

Ramendra K Sahoo

Xiao Tao Chang

Robert E Walkup

Ching-Farn Eric Wu

Hui-Fang Wen

Terry Lee Nelms

David Joseph Klepacki

Simone Sbaraglia

David Allen Dennerline

David Paul LaPotin

Pavlos Maniotis

Xiaolu Mei

Claudia Misale

Ming-Hung Chen

Terry Lee Nelms Ii

Disclosed embodiments provide a virtual topology in response to an allocation request. The virtual topology is a synchronized subgraph of the physical topology. The synchronized subgraph virtual topology mirrors the physical topology in terms of tree structure. However, merely sharing the physical topology with users/customers is not feasible, as it may reveal additional infrastructure details such as MAC addresses, IP addresses, and the like, that can compromise security and/or jeopardize multi-tenant operations. Disclosed embodiments create virtual topology structures with obfuscated node data, so that important data such as physical IP addresses and/or MAC addresses are hidden from the end-users. Thus, disclosed embodiments provide the benefits of performance enhancement that comes from sharing the topology without the downside of compromising security by revealing physical topology details.

Resource topology generation for computer systems

Communication endpoints associated with a virtual machine for migration can be discovered. Properties associated with the communication endpoints can be discovered. The communication endpoints can be matched to establish a communication channel between the endpoints. The communication endpoints, the properties and the at least one communication channel can be used in recreating the communication channel for a migrated version of the virtual machine.

Discovery and recreation of communication endpoints in virtual machine migration

A server LPAR operating in a virtualized computer shares pages with client LPARs using a shared memory region (SMR). A virtualization function of the computer receives a get-page-ID request associated with a client LPAR to identify a physical page corresponding to a shared page included in the SMR. The virtualization function requests the server LPAR to provide an identity of the physical page. The virtualization function receives a page-ID response comprising the identity of a server LPAR logical page that corresponds to the physical page. The virtualization element determines a physical page identity and communicates the physical page identity to the client LPAR. The virtualization element receives a page ID enter request and enters an identity of the physical page into a translation element of the computer to associate a client LPAR logical page with the physical page.

Shared memory in a virtual environment

A server logical partition (LPAR) of a virtualized computer includes shared memory regions (SMRs). The SMRs include pages of the server LPAR memory to share with client LPARs. A hypervisor utilizes an export vector to associate logical pages of the server LPAR with SMRs. The hypervisor further utilizes a reference array to associate SMRs with client LPARs that have mapped at least one physical memory page of the SMR from a logical page of the client LPAR memory. In processing an operation to unmap one or more shared physical pages from one or more LPARs, the hypervisor uses the export vector and reference array to determine which LPARs have had a mapping to the physical pages.

Invalidation of shared memory in a virtual environment

A graph processing system includes a graph API (Application Programming Interface), as executed by a processor on a computer, and that includes a plurality of graph operators to create graphs and to execute graph analytic applications on the created graphs, the graph operators supporting a creation and manipulation of multi-dimensional properties of graphs. A run-time system is executed by the processor and implements routines that dynamically adjust a plurality of representations and algorithms to execute sequences of operations on graph data. A library is accessible to the run-time system and stores a specification of calling signatures for the graph operators such that the graph operators can be called from any of various computer programming languages such that top-level algorithms received in an input graph application can be understood in the graph processing system when received in any of the various computer programming languages. Thereby the top-level algorithms written to the graph API are portable across multiple implementations.

Sophisticated run-time system for graph processing

A graph processing system includes a graph API (Application Program Interface), as executed on a processor of a computer and as capable of implementing any of a plurality of graph operators to express computations of input graph analytics applications. A run-time system, executed by the processor, implements graph operators specified by each graph API function and deploys the implemented graph operators to a selected computing system. A library contains multiple implementations for each graph API function, each implementation predetermined as being optimal for a specific set of conditions met by a graph being processed, for functional capabilities of a specific computing system on which the graph is being processed, and for resources available on that specific computing system.

Optimized system for analytics (graphs and sparse matrices) operations

Techniques for replacing one or more blocks in a cache, the one or more blocks being associated with a plurality of data streams, are provided. The one or more blocks in the cache are grouped into one or more groups, each corresponding to one of the plurality of data streams. One or more incoming blocks are received. To free space, the one or more blocks of the one or more groups in the cache are invalidated in accordance with at least one of an inactivity of a given data stream corresponding to the one or more groups and a length of the one or more groups. The one or more incoming blocks are stored in the cache. A number of data streams maintained within the cache is maximized.

Maintaining a cache of blocks from a plurality of data streams

This invention provides a request controlling apparatus, processor and method. The request controlling apparatus is connected to a request storage unit and includes: a queue unit storing flag recording region configured to record a storing flag corresponding to a queue unit in the request storage unit, a comparing means configured to judge whether a incoming first queue unit corresponds to a same message as an already existing queue unit, where the already existing queue unit is in the request storage unit and a flag setting means is configured to set the storing flag corresponding to the already existing queue unit in the queue unit storing flag recording region, to indicate that a message state related to the already existing queue unit will not be stored if the first queue unit corresponds to the same message as in the already existing queue unit.

Request controlling

Improved techniques are disclosed for use in an intrusion prevention system or the like. For example, a method comprises the following steps performed by a computing element of a network. A packet of a flow is received, the flow comprising a plurality of packets, wherein the plurality of packets represents data in the network. A network intrusion analysis cost-benefit value is determined representing a benefit for analyzing the received packet for intrusions in relation to a cost for analyzing the received packet for intrusions. The method compares the network intrusion analysis cost-benefit value to a network intrusion analysis cost-benefit threshold to determine whether analyzing the received packet for intrusions before forwarding the received packet is warranted. Responsive to a determination that analyzing the received packet for intrusions before forwarding the received packet is not warranted, the received packet is forwarded, an indication is made that subsequent packets of the flow should be forwarded, and a determination is made whether the received packet indicates an intrusion after forwarding the received packet.

Network intrusion protection

A method for application performance data collection includes steps or acts of: customizing a performance tool for collecting application performance data of an application; modifying the application by inserting the performance tool while the application does not need to be rebuilt from the source; executing the application; and collecting the application execution performance data such that only interesting data is collected. Customizing the performance tool proceeds by implementing at least one configurable tracing function that can be programmed by the user; compiling the function(s) into an object file; and inserting the object file into the performance tool using binary instrumentation.

Binary programmable method for application performance data collection

A method for decompressing multiple data streams includes receiving a packet of data of a compressed data stream, directing the received packet to a selected one of a plurality of decompression functional units within a hardware-based decompression accelerator unit, obtaining decompression state information pertaining to the compressed data stream, and decompressing the received packet using the obtained decompression state information.

Method and apparatus for concurrent and stateful decompression of multiple compressed data streams

Growing community of inventors

Valhalla, NY, United States of America

Hao Yu