Abstract: An open domain question answering apparatus includes a communication parser and a machine learning module. The communication parser parses a communication from a user account. The communication parser uses a natural language interpretation algorithm to construct a query and identify a product. The machine learning module is coupled to the communication parser and populates a plurality of potential responses to the query based on data associated with the identified product. Each of the plurality of potential responses has an associated level of detail. The machine learning module applies a machine learning user capacity model to the plurality of potential responses to identify a response with an associated level of detail commensurate with the machine learning user capacity model. The machine learning user capacity model is trained using data relating to a cognitive capacity of the user account.

Abstract: In remote power down control of a device configured for externally initiated remote power down, such as over a network, all externally connected sources of valid power down requests for the device is monitored; and the device is powered down in response to a combination of either of a received power down request and a time out with respect to each of the sources of valid power down requests for the device.

Abstract: A method for selectively compressing unstructured data includes preparing, at a host system, a data blob made up of unstructured data, for storage on a storage system. The method encodes a first portion of the data blob with a first color and a second portion of the data blob with a second color. The first color indicates that the associated portion should remain uncompressed, while the second color indicates that the associated portion should be compressed. The method transmits the encoded data blob to a storage virtualization layer associated with the storage system. In response, the storage virtualization layer stores the first portion in an uncompressed format, and compresses the second portion and stores the second portion in a compressed format. A corresponding system and computer program product are also disclosed herein.

Abstract: An apparatus, system, and method are disclosed for handling data being communicated over lossless Ethernet that is sensitive to delays. Fiber Channel over Ethernet (FCoE) is one example of an environment where data may be subject to unacceptable delays. The method involves designating certain data as low latency data that is sensitive to delays in transmission. The low latency data is then transmitted in such a manner that the receiving devices are aware that they are receiving low latency data. If a delay in the transmission of low latency is detected, commands are issued that pause or slow standard data in order to free up bandwidth for the low latency data. The commands may be, for example, backward congestion notifications and priority flow control. Low latency data is exempted from backward congestion notifications and priority flow control. Priority 7 priority group 15 may be reserved exclusively for low latency data.

Abstract: The present invention relates to a method for data access in a storage infrastructure. The storage infrastructure comprises a host system connected to at least a first storage system and a second storage system. The first storage system receives, from the host system, a write request for storing a data chunk, the write request is indicative of a first identifier of the data chunk. The first storage system calculates a hash value of the received data chunk using a hash function. The first storage system determines a first storage location in the first storage system of the data chunk and sends a write message including the hash value, the first identifier and the first storage location to the de-duplication module. The de-duplication module determines whether the hash value exists in the data structure.

Abstract: A cognitive communication assistant receives a message transmitted over a communication network from a sender to a recipient. A sender's industry identified with the sender and a recipient's industry identified with the recipient are determined. One or more terms associated with the sender's industry are extracted from the message. A definition associated with the one or more terms is searched for in an on-line reference text. The message is updated based on the definition. The message is transmitted over the communication network to the recipient.

Abstract: Various embodiments for regulating link speed for performance enhancement and port diagnosis are provided. In response to identifying an amount of errors in a communications link above a predetermined threshold, an applicable transmission speed is selectively reduced. If errors identified at the reduced transmission speed are found to decrease, a communications port incorporating the communications link is flagged as potentially dirty, and if the errors identified at the reduced transmission speed are found to remain constant, the communications port is flagged as potentially bad.

Abstract: According to one embodiment, an independent file server connected to a wide area network (WAN) includes local file storage, a file system comprising information about directory structures, path, and file names of other file systems of other file servers connected to the WAN, and a deduplication module configured to communicate with other deduplication modules of other file servers connected to the WAN, perform data deduplication of data objects stored locally in the local file storage, and create and maintain a deduplication table mapping data chunks of locally stored data objects to identification characteristics and storage location in the local file storage, and construction plans of data files including locations of data chunks within the data files. Other systems, methods, and computer program products are presented according to more embodiments.

Abstract: Various embodiments for managing data integrity in a computing storage environment, by a processor device, are provided. In one embodiment, a method comprises applying a copy-on-write technique to a cache in a computer storage system such that each write arriving in the cache is assigned to a separate new physical location and registered sequentially in an order the write arrived, for preserving a state of the computer storage system during a failure event.

Abstract: Various embodiments for managing data transmission errors in a computing storage environment, by a processor device, are provided. In one embodiment, a method comprises, in response to reaching a predetermined error threshold while routing data traffic through the computing storage environment, wirelessly transmitting at least certain data traffic from a source to a switch nearest a destination of the certain data traffic.

Abstract: According to one embodiment, a calibration assembly includes an outer layer having at least one calibration trench extending along a y-axis, and an encapsulation layer within the calibration trench. The encapsulation layer has a plurality of nanoparticles spaced apart along said y-axis of said at least one calibration trench. Each of said plurality of nanoparticles are provided at known y-axis locations in said calibration trench, and each of the plurality of nanoparticles have a known magnetic property.

Abstract: In one embodiment, a method includes creating a cartridge memory (CM) database accessible via a storage cloud, the CM database storing information corresponding to a plurality of data storage cartridges that are configured to store data that is accessible by one or more data storage drives located in one or more data storage libraries. The method also includes storing information corresponding to each of the plurality of data storage cartridges to separate entries in the CM database, the information being selected from a group including: a history of read errors for each data storage cartridge that occurred during reading of the data storage cartridge, which data storage drive was used for writing data to each data storage cartridge, and a temperature and humidity within each data storage library, within each data storage cartridge, and within a data storage drive at a time when a data storage cartridge was written.

Abstract: Methods and systems are provided for displaying information on a hard disk drive assembly. One hard disk drive assembly includes a hard disk drive that includes a driver circuit coupled to a display and to a display memory. The driver circuit activates the display mounted on the hard disk drive assembly. In response to the activation of the display the driver circuit obtains information from the display memory of the hard disk drive. The driver circuit displays at least a portion of the information obtained from the display memory on the display mounted on the hard disk drive assembly.

Abstract: An apparatus for managing network security includes an inventory module, a survey module, an observation module, and a security module. The inventory module identifies each node of a network. The survey module gathers information for each node of the network. The security module generates notifications to one or more users in response to the observation module detecting one or more irregularities.

Abstract: In remote power down control of a device configured for externally initiated remote power down, such as over a network, all externally connected sources of valid power down requests for the device is monitored; and the device is powered down in response to a combination of either of a received power down request and a time out with respect to each of the sources of valid power down requests for the device.

Abstract: In one embodiment, a system includes a processor and logic integrated with and/or executable by the processor, the logic being configured to cause the processor to determine that a link configured to transmit data between two end points in a networking system has become degraded, encode data prior to transmitting the data across the degraded link, and transmit the encoded data across the degraded link provided that data transmitted across a normal link is not encoded. In another embodiment, a method for selectively overcoming a degraded link in a networking system includes determining that a link configured to transmit data between two end points has become degraded, encoding data prior to transmitting the data across the degraded link, and transmitting the encoded data across the degraded link.

Abstract: In one embodiment, a system includes a storage cloud comprising storage media, a cartridge memory (CM) database stored to the storage media, and a storage controller, wherein the storage controller is configured to communicate with one or more data storage drives located in one or more data storage libraries, each data storage drive being configured to write and/or read data to/from a plurality of data storage cartridges therein, and control the one or more data storage libraries as a single logical library, wherein the CM database includes a plurality of entries, each entry corresponding to a data storage cartridge. In another embodiment, the data storage cartridge includes a data storage medium configured to store more than about 100 MB of data, a physical CM configured to store less than about 100 KB of data, and a visible indicator displayed on an exterior of the housing.

Abstract: Simulating latency in a network environment. A device sends a latency request. The device receives a latency support confirmation. The device builds an I/O frame. The I/O frame comprises a latency simulating bit and a latency duration. Based on the latency simulating bit and the latency duration, the device holds the I/O frame. The device sends the I/O frame.

Abstract: Simulating latency in a network environment. A first device receives a latency request. The first device sends a latency support confirmation. The first device receives an I/O frame. The I/O frame comprising a latency simulating bit and a latency duration. Based on the latency simulating bit and the latency duration, holding, by the first device, the I/O frame. The first device sends the I/O frame on.

Abstract: In one general embodiment, a method for determining a controlling entity of a first avatar in a virtual world includes: receiving a request for access to a resource via a first avatar; providing a challenge to the first avatar in response to receiving the request; receiving a challenge response via the first avatar in response to the challenge; determining an identity confidence level based on the challenge response; making an identity determination for a controlling entity of the first avatar based on the identity confidence level; and providing or denying access to the resource based on the identity determination.