Abstract: In one embodiment, a computer system includes a first client subsystem, a second client subsystem and a network. The network is coupled to convey a packet between the first client subsystem and the second client subsystem. The network is partitioned into a plurality of network slices and the packet is partitioned into a plurality of packet slices. Each packet slice is conveyed on a respective one of the plurality of network slices. At least one of the plurality of packet slices includes redundant information associated with the packet and is usable to regenerate at least another of the plurality of packet slices. In addition, each of the plurality of network slices corresponds to a field replaceable hardware unit.

Abstract: A trace cache for efficient self-modifying code processing enables selective invalidation of entries of the trace cache, advantageously retaining some of the entries in the trace cache even during self-modifying code events. Instructions underlying trace cache entries are monitored for modification in groups, enabling advantageously reduced hardware. Associated with trace cache entries are one or more translation ages, determined when the entry is built by sampling current ages of memory blocks underlying the entry. When the entry is accessed and micro-operations therein are processed, each of the translation ages of the accessed entry are compared with the current ages of the memory blocks underlying the accessed entry. If any of the age comparisons fail, then the micro-operations are aborted and the entry is invalidated. When any portion of a memory block is modified, the current age of the modified memory block is incremented.

Abstract: Disclosed is a visual method of communicating an abnormal situation (e.g., the crashed or downed state) of the computer system including a display. The system status is thus indicated to the user. Upon a system crash or downed state, the display entirely or in part (a first area) is proportionally decreased in brightness. Brightness of a second area (which may overlap the first area) is increased in full or in part by a fixed amount. This results in a translucent film overlay which preserves in the background the display screen content existing at the moment the system crashed. The translucent film overlay may have predefined characteristics, such as tint to signify system status and/or level or degree of the system's problem.

Abstract: A node in a multi-node system includes a memory, an active device that includes a cache, an interface that sends and receives coherency messages on an inter-node network coupling the node to another node, and an address network that communicates address packets between the devices in the node. In response to receiving a coherency message from the other node requesting an access right to a coherency unit, the interface sends an address packet on the address network. The address packet is a first type of address packet if the coherency unit is in the modified global access state in the node and a second type of address packet otherwise. If the active device is the owner of the coherency unit, the active device responds to the first type of address packet and ignores the second type of address packet.

Abstract: A small footprint device can securely run multiple programs from unrelated vendors by the inclusion of a context barrier isolating the execution of the programs. The context barrier performs security checks to see that principal and object are within the same namespace or memory space and to see that a requested action is appropriate for an object to be operated upon. Each program or set of programs runs in a separate context. Access from one program to another program across the context barrier can be achieved under controlled circumstances by using an entry point object. The entry point object can either perform accesses across the context barrier on behalf of a requesting program or can pass the request to the program to be accessed and switch contexts to the program to be accessed.

Abstract: Encrypted data packets are received by a network interface card. The network interface card, upon determining that the received data packets are encrypted, directs the encrypted data packets to decryption hardware in the network interface card. The decryption hardware decrypts the encrypted data packets and forwards the decrypted data packets to a hardware classifier that classifies the decrypted data packets and directs the classified decrypted data packets to the appropriate receive resource(s) of the network interface card.

Abstract: A method for configuring a switch with a switch controller that includes obtaining a switch program, wherein the switch program comprises a plurality of instructions written using a switch control language, and executing the switch program on the switch controller, wherein executing the switch program includes for each instruction in the plurality of instructions determining a switch specified by the instruction, sending the instruction to the switch, and configuring the switch based on the instruction.

Abstract: In a generational garbage collector, a decision is made to pre-tenure, or allocate new objects directly in the old generation, by a two step process. In the first step, during a young-generation collection, the number of bytes that survive collection is determined for each allocation site and a predetermined number of sites with the highest number of surviving bytes are selected as candidate sites. In the second step, during a subsequent young-generation collection, the survival rates are determined for the candidate sites and objects to be allocated from sites with a sufficiently high survival rate are allocated directly in older generations.

Abstract: In one embodiment, a method and apparatus for predicting and preventing escalations is disclosed. The method includes generating a predictive model from analysis of historical defect data and corresponding historical escalation data, and applying the predictive model to current defect data to determine an escalation risk of the current defect data.

Abstract: A system for evaluating a simulation includes a reference simulator configured to execute a simulation image to obtain golden data, a test simulator configured to execute the simulation image to obtain test data, and a comparator configured to generate a comparison result by comparing a portion of the golden data to a portion of the test data before the execution of the simulation image on the test simulator has completed.

Abstract: A composite interconnect system includes a plurality of carbon nanotubes, a plurality of solder balls and standoff balls disposed on a first device to provide a connection to a second device. A die-attached substrate includes a substrate and one or more die disposed on the substrate by a die-attach composite interconnect. The die-attach composite interconnect includes a plurality of carbon nanotubes, solder bumps, and standoff balls disposed on the die to provide one or more connections to the substrate. A PCB-attached substrate package includes a substrate package and one or more die disposed on the substrate package. The substrate package is disposed on a PCB by a PCB-attach composite interconnect. The PCB-attach composite interconnect includes a plurality of carbon nanotubes, solder balls, and standoff balls disposed on the substrate package to provide one or more connections to the PCB.

Abstract: A computer program for software difference comparison is provided. The program extracts data from the files on the hard disk, including data such as symbols extracted from symbol tables, APIs extracted from help files, and/or configuration information. This information may be collected at two or more different times, for example, before and after a version of software is updated to a new version of the software. The collected data is extracted into a relational database. The relational database may be used to determine the differences between multiple versions of software, or between one piece of software and another.

Abstract: A method for storing a data block, involving storing the data block in a storage pool, obtaining a data block location, calculating a data block checksum for the data block, and storing a first indirect block in the storage pool, wherein the first indirect block comprises the data block location and the data block checksum.

Abstract: One embodiment of the invention provides a method of operating a switch with cut-through routing in a communications network. The method includes receiving a packet at a first port of the switch and determining a second port of the switch from which the received packet is to be forwarded out over the network. At least one authentication check is performed on the received packet. In general, the switch starts to forward the received packet from the second port prior to complete reception of the packet at the first port. However, this onward transmission is delayed until the authentication check has completed. This ensure that none of the received packet is forwarded from the second port if any authentication check fails.

Abstract: A virtual zone management framework manages one or more virtual computing zones forming a global computing zone residing on a host computer. The virtual zone manager comprises a normalization engine, a zone probe, and an interface engine that manages each virtual computing zone residing on the host computer. The virtual zone manager manages the services operating on each virtual zone while simultaneously managing the global computing environment using probes and/or agents to gather data and zone characteristics in response to a triggering event. Once data of interest are gathered for each zone, they are normalized so as to present a consistent and compatible representation of the characteristics with respect to the other virtual computing zones. Based on a comparison and analysis of the normalized characteristics, the virtual zone manager manages each of the virtual zones within the global computing environment as well as the global computing environment itself.

Abstract: A method to instrument software code may comprise marking one or more blocks of code in an application program to provide a marked code. The marking may comprise inserting a marker at each of the one or more blocks according to a marking application programming interface (API). The one or more of the blocks of code marked according to the marking API may be transformed so that one or more functions are performed for the marked blocks. The one or more functions may not be part of the application program. The method may include executing the application program and performing the one or more functions for the marked blocks.

Abstract: Systems, methods, and products provided for facilitating the reuse of data blocks. In one implementation, a registration server receives request data objects from programs that request use of a data block. When the data block is not already registered with the registration server, the registration server registers the data block, for example in a list of registration data objects. A registration handle object is returned to the requesting program and provides a reference to the resources for accessing the data block. The present invention facilitates efficient use of resources in connection with multiple requests for a data block because resources needed for handling the binary data block are not duplicated.

Abstract: One embodiment of the present invention provides a system that facilitates loading of an in-place executable file into a virtual machine. The system operates by loading the in-place executable file into the virtual machine, where the in-place executable file contains internal pointers that point to memory locations within the in-place executable file, and a target memory address that specifies a target location in memory where the in-place executable file is configured to be loaded. The system then determines an actual memory address at which the in-place executable file is actually loaded. If the actual memory address and the target memory address are different, the system generates an offset value, which is the difference between the actual memory address and the target memory address. Finally, the system adjusts each internal pointer within the in-place executable file by the offset value so that the internal pointers point to intended locations inside the in-place executable file.

Abstract: A method for assigning work to a plurality of threads using a primitive data element to partition a work load into a plurality of partitions. A first partition is assigned to a first thread and a second partition is assigned to a second thread of the plurality of threads. A method for improving the concurrency of a multithreaded program by replacing a queue structure storing a plurality of tasks to be performed by a plurality of threads with a partition function. A computer system including a processor unit configured to run a plurality of threads and a system memory coupled to the processor unit that stores a multithreaded program. The multithreaded program workload is partitioned into a plurality of partitions using a primitive data element and a first partition of the plurality of partitions is assigned to a first thread of the plurality of threads for execution.

Abstract: A method and apparatus for equalizing a reflection in a reflective high speed serial link. The method involves obtaining an amplitude and delay time of a compensating pulse that is transmitted in response to a pulse transmitted on the serial link. The apparatus comprises a programmable delay element and a driver stage configured to transmit a delayed and amplitude adjusted version of a pulse transmitted on the serial link. A method for equalizing a plurality of reflections in a reflective high speed serial link. The method involves obtaining an amplitude and delay time of a first compensating pulse and an amplitude and delay time of a second compensating pulse. The method further involves transmitting the first compensating and second compensating pulses in response to a pulse transmitted on the serial link.