Sun Microsystems, Inc.

Intel Corporation

Andrew M Rudoff

Emrys J Williams

Wei P Chen

Rajat Agarwal

Michael W Shapiro

Timothy Patrick Haley

Cynthia A McGuire

Kuljit Singh Bains

Andrew V Anderson

Wei Wu

Alaa R Alameldeen

Mark Maybee

Theodros Yigzaw

Zhe Wang

Jing Ling

Deep K Buch

John G Holm

Sreenivas Mandava

Kjersten E Criss

Mark Shellenbaum

Neil Veness Perrin

Hsing-Min Chen

Matthew T Simmons

Vaibhav Singh

A system can dynamically migrate memory pages from near memory to far memory during runtime. A system basic input output system (BIOS) can program a first memory address space of size P and a second memory address space of size P to a near memory (NM) space of size (N) and a far memory (FM) space of size (M), where P equals N+M. For the first memory address space, the OS can manage the NM space and the FM space as a flat memory space with an address space of size P available. For the second memory address space, the OS can manage the NM space as a NM cache for FM, with an address space of size M available.

Dynamic multilevel memory system

A memory subsystem includes memory devices with space dynamically allocated for improvement of reliability, availability, and serviceability (RAS) in the system. Error checking and correction (ECC) logic detects an error in all or a portion of a memory device. In response to error detection, the system can dynamically perform one or more of: allocate active memory device space for sparing to spare a failed memory segment; write a poison pattern into a failed cacheline to mark it as failed; perform permanent fault detection (PFD) and adjust application of ECC based on PFD detection; or, spare only a portion of a device and leave another portion active, including adjusting ECC based on the spared portion. The error detection can be based on bits of an ECC device, and error correction based on those bits and additional bits stored on the data devices.

Adaptive error correction to improve system memory reliability, availability, and serviceability (RAS)

Examples herein relate to checkpoint replication and copying of updated checkpoint data. For example, a memory controller coupled to a memory can receive a write request with an associated address to write or update checkpoint data and track updates to checkpoint data based on at least two levels of memory region sizes. A first level is associated with a larger memory region size than a memory region size associated with the second level. In some examples, the first level is a cache-line memory region size and the second level is a page memory region size. Updates to the checkpoint data can be tracked at the second level unless an update was previously tracked at the first level. Reduced amounts of updated checkpoint data can be transmitted during a checkpoint replication by using multiple region size trackers.

Virtual machine replication and migration

One embodiment of the invention provides apparatus including a data structure representing a fault tree for a system. The data structure comprises a plurality of events linked by propagations. Each event is classified as one of at least three possible event types. A first type of event is a problem event, which represents an underlying cause of misbehavior in the system. A second type of event is an error event, which represents an error in the system comprising an incorrect signal or datum. A third type of event is a report event, representing the formal detection by the system of an error. Each propagation in the fault tree denotes a cause and effect linkage from one event to another event. There are no propagations within the fault tree to a problem event.

System and method for providing a data structure representative of a fault tree

A method and system are provided for automated diagnosis for a system. In one embodiment, the method includes providing a fault tree representation of the system, the fault tree specifying propagations of errors generated in the system by problems to produce error reports. At least some of the propagations have timing information associated therewith. One or more error reports having timing information associated therewith are received and analysed using the fault tree representation to determine a suspect list of problems. The suspect list contains those problems that could have generated errors to produce the received error reports compatible with the propagations in the fault tree, and consistent with the timing information associated with the propagations and the received error reports.

System and method for automated problem diagnosis

A method, apparatus, and computer program product diagnosing and resolving faults is disclosed. A disclosed fault management architecture includes a fault manager suitable having diagnostic engines and fault correction agents. The diagnostic engines receive error information and identify associated fault possibilities. The fault possibility information is passed to fault correction agents, which diagnose and resolve the associated faults. The architecture uses logs to track the status of error information, the status of fault management exercises, and the fault status of system resources. Additionally, a soft error rate discriminator can be employed to track and resolve soft (correctible) errors in the system. The architecture is extensible allowing additional diagnostic engines and agents to be plugged in to the architecture without interrupting the normal operational flow of the computer system.

Error reporting to diagnostic engines based on their diagnostic capabilities

A method for managing a fault involves detecting an error, gathering data associated with the error to generate an error event, and categorizing the error event using a hierarchical organization of the error event.

Event protocol and resource naming scheme

One embodiment of the invention provides a method for simulating the operation of a system. The method includes providing a fault tree representation of the system. The fault tree defines a set of problems that may occur in the system, and specifies propagations in the system whereby a problem may create one or more errors that may in turn be detected by error detectors to produce corresponding error reports. The fault tree representation allows the presence of a problem in the system to be simulated, and the set of error reports resulting from the simulated problem to be determined. This simulation can be repeated for different problems to compare the sets of error reports potentially produced by the different problems.

System and method for simulating system operation

A method and system are provided for generating a data structure representative of a fault tree for a system. One embodiment of the method includes providing one or more input files comprising a source code description of the fault tree. The source code description has a plurality of statements specifying events and propagations in the fault tree, where a propagation represents a cause and effect linkage between events. The method further includes compiling the input files in source code into the data structure.

System and method for generating a data structure representative of a fault tree

A mechanism for replicating and maintaining files in a space-efficient manner is disclosed. In one embodiment, a current file is replicated by associating the data blocks of the current file with both the current file and a new file. The new file and the current files are separate and distinct from each other. Data blocks remain shared until the content of one file is modified. Only the data blocks affected by the modification are duplicated. The two files share the unchanged data blocks, but both files are each associated with additional data blocks that reflect the modification. Reference values are used to track how many files are associated with each data block. In another embodiment, a file descriptor is linked to a filename at the request of a user or application. The present invention avoids needless duplication of the same data blocks among copies of a file.

Mechanism for replicating and maintaining files in a space-efficient manner

A method and system are provided for performing automated management on a system. The method includes providing a fault tree representation of the system. The fault tree specifies propagations of errors generated in the system by problems to produce error reports. The method further includes receiving one or more error reports which are analyzed using the fault tree representation to determine one or more suspected problems. The suspected problems represent those problems that could have generated errors to produce the received error reports compatible with the propagations in the fault tree. The method further includes deconfiguring or reconfiguring one or more components in the system in response to the determination of the suspected problems.

System and method for performing automated system management

A system for providing extended file attributes is described. Logically, attributes are supported as files within the file system. In this way, the file system is augmented with an orthogonal namespace of file attributes. According, any file (including attribute files) may have an arbitrarily deep attribute tree associated with it. Access to the attribute values is via file descriptors obtained via a special attribute interface. This logical view of &ldquo;attributes as files&rdquo; allows the leveraging of existing file system interface functionality to support the construction, deletion, and manipulation of attributes.

Method and apparatus for providing extended file attributes in an extended attribute namespace

Mechanism for replicating and maintaining files in a spaced-efficient manner

Improved techniques for providing users and application programs with access to data stored in computer systems are disclosed. The improved techniques are particularly well suited for use in computer systems where data can be logically organized in files presented in a file system. A multilevel caching system suitable for storing information relating to files in the file system is provided. The stored information can include file references suitable for locating files in the file system as well as other useful information about the file system. The multilevel caching system provides the ability to implement various caching strategies at different levels and increases the probability of cache hits when seeking to locate files in a file system. Accordingly, relatively expensive read operations to persistent storage devices can be minimized when locating files in the file system.

Method and apparatus for locating data in computer systems

A bootstrap system for a computer with peripheral devices has a first level boot firmware program and a second level boot program loaded from disk by the first level program. The second level boot program contains a device tree constructed according to IEEE-1275 (Open Firmware) standards, describing currently known peripheral devices and computer resource allocations. The device tree is modifiable during boot operations to incorporate information about additional devices. The second level boot program scans I/O buses for additional devices not included in the current device tree and loads drivers only for the additional devices. The second level boot program further allows a user to enter information into the device tree. The second level boot program is preferably implemented under DOS emulation, and may be developed in a DOS-based code development environment and ported to the computer.

Growing community of inventors

Boulder, CO, United States of America

Andrew M Rudoff