News Column

Patent Issued for Condition Detection and Reporting in Complex Systems

February 13, 2014



By a News Reporter-Staff News Editor at Computer Weekly News -- From Alexandria, Virginia, VerticalNews journalists report that a patent by the inventor White, Philip (San Mateo, CA), filed on February 17, 2011, was published online on January 28, 2014.

The patent's assignee for patent number 8639971 is Scale Computing (San Mateo, CA).

News editors obtained the following quote from the background information supplied by the inventors: "In complex systems, particularly in relatively complex data storage and retrieval systems, it might occur that conditions of individual components of the system should be noticed, and action taken thereon. For just one example, in a RAID storage system it might occur that one of the disk drives becomes unavailable or unreliable. In such cases, the system might respond by alerting an operator to that condition, and upon re-availability or the presence of a newly available and reliable disk drive, the system might take remedial measures, such as for example re-striping and recomputing parity for a set of striped data blocks formerly maintained, in whole or in part, on the disk drive that was repaired or replaced.

"One problem in the known art is that when such systems are relatively complex, whether in design, in implementation, or in operation, action taken by one component of the system might have an unexpected effect upon other components of the system. For just one example, in a storage system with multiple tiers, such as for example a system including paging memory blocks to storage on disk, it might occur that one element of the system might be attempting to read data from, or write data to, a disk drive, while another element of the system is attempting to determine if that disk drive is reliable or if its data should be moved to another location. Multiple such operations might be inconsistent if performed concurrently, or at least without provision for the possibility that such multiple operations might occur.

"Known methods include operating system locking techniques. While these techniques generally achieve their purpose of preventing inconsistent operations from occurring concurrently, they have at least the drawback that they involve careful checks, either during design and operation, to assure that they are uniformly consistent and do not interfere."

As a supplement to the background information on this patent, VerticalNews correspondents also obtained the inventor's summary information for this patent: "This description includes techniques, including methods, physical articles, and systems, which provide the effect of maintaining consistency and freshness of information about a system in operation, and which provide the effect of maintaining consistency of actions by system actors. Consistency and freshness of information about a system might include hardware consistency, that is, information about the status of hardware elements of the system, software consistency, that is information about the status of software elements of the system, and otherwise, and combinations and conjunctions thereof.

"Consistency of information has the effect of assuring that all elements that rely on status of the system will operate using only a single global status of the system for any particular status time. For example and without limitation, no two elements will ever act upon status values that are inconsistent. Freshness of information has the effect of assuring that each element that relies on status values of the system will operate only with status values that have been measured sufficiently recently as to be reliable. Consistency of actions has the effect of assuring that all elements that act in reliance upon status of the system will operate pairwise consistently. For example and without limitation, no two elements would operate at cross-purposes to one another.

"While this application describes errors and inconsistencies relating to hardware types of failures, such as for example disk related errors, in the context of the invention, there is no particular requirement for any such limitation. For example and without limitation, it might sometimes occur that errors, failures, or inconsistencies might appear or occur in response to actions by software components, particularly but not limited to those software components integrated into portions of the operating system or firmware by another vendor, obtained from external vendors, or otherwise, or some combination or conjunction thereof.

"In one embodiment, it might occur that software errors, also known as bugs, might appear or occur in response to third-party components, the interaction of more than one such third-party components, the interaction of such third-party components with trusted code, or otherwise, or some combination or conjunction thereof. Accordingly, the reader is urged not to limit the description of the invention as relating only to hardware types of failures, but also to understand that software types of failures, or combinations or conjunctions of hardware and software failures, such as for example relatively unknown software bugs arising out of operation of third-party components.

"For example and without limitation, external software components might include information collectors that generally provide useful data, but which infrequently or otherwise from time to time erroneously provide undocumented or unexpected output values (such as for example, floating-point values where integer values are expected), or which erroneously provide output values that are clearly invalid (such as for example, negative values for available disk free space or disk size), or which generate crash results, invalid interrupts, memory leaks or other memory errors, or other types of software-related (or combinations or conjunctions of software-related and hardware-related) errors. In such cases, an information container's metadata might indicate that the external software component has generated an error, might indicate a type of failure, a specific type of failure, a possible corrective action or possible likely non-erroneous value, or a component, sub-component, or other logic associated with a decision function that might be able to respond to such metadata.

"This has the effect that consistency and freshness of information, or the single global status of the system, have the advantage of insulating trusted code from otherwise unapparent or unknown errors in hardware or software components from external vendors. Sometimes those hardware or software components from external vendors are relied upon for accurate and reliable operation of a system; in such cases, it is desirable that otherwise unapparent or unknown errors from external vendors be identified and, preferably, ameliorated, without substantial negative system problems.

"These features might be provided in combination or conjunction with techniques which include one or more of the following:

"a substantially independent level or set of information collectors, each of which is responsive to a status aspect of the system; a substantially independent level or set of information containers, each of which is responsive to one or more information collectors, and each of which maintains both status values and metadata about those status values indicative of whether those status values can be reliably used by system elements, and of whether those status values are consistent with known correct values (that is, are consistent with being accurate); a substantially independent level or set of information conditions, each of which is responsive to one or more information containers or other information conditions, and each of which is logically consistent with a single global status of the system; and a substantially independent level or set of system actors, each of which is responsive to one or more information containers or information conditions, and each of which has an associated action, those actions being logically consistent with a single global status of the system.

"In one embodiment, system actors might be divided into (1) notifiers, which might be read-only with respect to the system and serve to present notifications to system monitors or system operators, and (2) ameliorators, which might both read system status and act upon that system status to correct errors, distribute or redistribute load, or otherwise detect and correct system operation gone awry."

For additional information on this patent, see: White, Philip. Condition Detection and Reporting in Complex Systems. U.S. Patent Number 8639971, filed February 17, 2011, and published online on January 28, 2014. Patent URL: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&p=15&u=%2Fnetahtml%2FPTO%2Fsearch-bool.html&r=712&f=G&l=50&co1=AND&d=PTXT&s1=20140128.PD.&OS=ISD/20140128&RS=ISD/20140128

Keywords for this news article include: Software, Scale Computing.

Our reports deliver fact-based news of research and discoveries from around the world. Copyright 2014, NewsRx LLC


For more stories covering the world of technology, please see HispanicBusiness' Tech Channel



Source: Computer Weekly News


Story Tools