[go: up one dir, main page]

WO2000031639A1 - Element de controle d'un systeme de traitement de donnees - Google Patents

Element de controle d'un systeme de traitement de donnees Download PDF

Info

Publication number
WO2000031639A1
WO2000031639A1 PCT/EP1999/009055 EP9909055W WO0031639A1 WO 2000031639 A1 WO2000031639 A1 WO 2000031639A1 EP 9909055 W EP9909055 W EP 9909055W WO 0031639 A1 WO0031639 A1 WO 0031639A1
Authority
WO
WIPO (PCT)
Prior art keywords
applications
aem
availability
component
compliant
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/EP1999/009055
Other languages
German (de)
English (en)
Inventor
Markus Lautenbacher
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Siemens AG
Siemens Corp
Original Assignee
Siemens AG
Siemens Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Siemens AG, Siemens Corp filed Critical Siemens AG
Priority to EP99959313A priority Critical patent/EP1133729A1/fr
Publication of WO2000031639A1 publication Critical patent/WO2000031639A1/fr
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00—Error detection; Error correction; Monitoring
    • G06F11/008—Reliability or availability analysis

Definitions

  • Tele k ommunikationsdienste eg service nodes
  • Mission Critical Applications eg transactions in the financial sector
  • multimedia, interactive quasi-real-time network services it is necessary that the combination of computer platform (hardware, operating & Network Transport System [2] and network) and the applications running on it, i.e. a data processing system as a whole, offers particularly high availability.
  • the invention has for its object to overcome the disadvantages mentioned.
  • Availability Enhancing Middleware AEM offers a highly available service infrastructure as is necessary in areas such as telecommunications, finance, or interactive multimedia network services.
  • the AEM [3] shows the principle of how the AEM [3] fits into a computer system between the computer platform (Standard Operating & Network Transport System [2], network and hardware [1]) and the applications [4].
  • the A EM [3] is a pure software solution that is introduced into the computer as a new middleware layer between the standard operating & network transport system [2] including hardware [1] and the applications [4].
  • the middleware layer approach for the software implementation of the AEM allows the AEM to be made an inherent part of a newly developed computer system, but also enables retrofitting into an existing computer system.
  • Corresponding communication channels (denoted by (1) - (7) in FIGS. 2 and 3) effectively push the AEM between the normally direct communication of components [1], [2] and [4] in FIG. 1.
  • the AEM [3] controls the interaction (interaction) of applications [4] with the Operating & Network Transport System [2], network and hardware [1], and corrects any actions of the applications [4] with the aim of increasing the availability of the entire computer system (hardware, network, operating system, applications).
  • the AEM is able to integrate already existing applications (so-called non-AEM-compliant applications) (especially those existing applications that are only available in binary form) and on the other hand it has its own application programming interface (API) for special applications tailored the benefits of the AEM approach
  • AEM-compliant applications are available to offer these applications optimal access to the possibilities of the AEM.
  • FIG. 2 schematically shows an embodiment of the AEM middleware approach from FIG. 1 to increase the availability of data processing systems.
  • AEM architecture elements are designated with [ l ] - [5], communication connections between these elements with (l) - (7).
  • Network Transport System [2], AEM [3], several AEM-compliant [4a] and several non-AEM-compliant applications [4b]) are carried out as follows:
  • AEM-compliant applications [4a] communicate (1) via an open API provided by AEM [3a].
  • the API mediates (2) as an interface between the AEM-compliant applications [4a] and the corresponding subsystem [3b] of the AEM [3].
  • the AEM as a whole [3] checks and evaluates the incoming from the AEM-compliant application [4a]
  • Transport systems [3] are again monitored in the associated AEM subsystem [3b] and passed on to the AEM-compliant application [4a] (6). If the AEM [3] detects conflicts or problems, this is also signaled back to the AEM-compliant application [4a] (6).
  • AEM-compliant applications cover the handling of such feedback internally via the software standard technology of a so-called "event dealer” and delay then, for example, a memory request corresponding to the availability are of the overall system [5], this allows again.
  • AEM [4b] Applications that do not conform to AEM [4b] do not use the detour via AEM [3], but instead access (4) the resources of the Operating & Network Transport System [2].
  • the corresponding AEM subsystem [3c] monitors these system calls (e.g. through the "trace” system routine of the UNIX operating system) and the system messages generated thereby (5).
  • Analogous to the AEM-compliant applications [4a] the information obtained in this way about the non-AEM-compliant applications [4b] is considered by AEM [3] with regard to possible conflicts with other applications and compatibility with the availability of the overall system [5]. checked as a whole. If the AEM [3] detects corresponding problems, an attempt is made to eliminate them by stopping or terminating (e.g. UNIX signal "STOP" or "KILL”) of the corresponding non-AEM-compliant application [4b] (7).
  • terminating e.g. UNIX signal "STOP" or "KILL
  • FIGS. 1 and 2 show a detailed based on FIGS. 1 and 2
  • FIG. 3 also executes the AEM API [3a], the internal architecture of the AEM subsystems [3b] and [3c] and the associated AEM internal communication (il) - (i4).
  • the AEM API [3a] offers an interface that is secured with regard to overall availability for requesting so-called passive objects (PO), file access, memory handling, network communication, etc.
  • PO passive objects
  • the API offers corresponding so-called "stubs" object-oriented programming.
  • the interfaces are standard O perating system services such as FTP, TELNET, ... implemented as passive objects within the AEM [3]).
  • the A EM [3] holds information about the current state of the overall system [5] in the following central units:
  • This unit manages the runtime environment of the PO. All AEM-compliant applications are built from PO individual components using object-oriented methods.
  • This unit has the task of being networked
  • Security management Data and program code of different applications within the overall system should be kept separate from one another and attacks should be prevented.
  • the security management unit takes on this task.
  • This unit is a database in which persistent and temporary system-relevant information is kept; this includes information about the current system configuration with regard to hardware and software, the maximum values of the available system resources, and the resource profiles that are permitted
  • Resource management Define resource requirements per application, as well as the current system information about active applications.
  • This unit is responsible for managing local resources.
  • Resource management includes the detection of resource misuse by individual applications and the optimization of competing resource requirements by different applications.
  • the monitored resources include in particular CPU, memory, hard disk allocation, network connections.
  • This status information is updated via the exchange (2), which takes place via the AEM API [3a] via (1) with the AEM-compliant applications [4a], and internally (il) via the monitor unit.
  • the monitor acts as a central collection and
  • the monitor (3) communicates with the Operating & Network Transport System [2] of the AEM subsystem [3b] responsible for AEM-compliant applications [4a]. The monitor can therefore forward the AEM-compliant applications [4a] regarding the system status to the above-mentioned central units.
  • the monitor also receives indirect (i2) status information via the sensor unit, which is located in the AEM subsystem [3c] responsible for non-AEM-compliant applications [4b]. The monitor also forwards this status information to the above-mentioned central units.
  • the sensor monitors system calls (eg through the "trace” system routine of the UNIX operating system) and system messages (5) generated thereby, which are generated by non-AEM-compliant applications [4b] with direct access (4) to the resources of the Operating & Network Transport System [2] can be generated.
  • Deviates the detected in the monitor actual value of G eticianstatus from an adjustable set-profile eg respect. The number of active applications, machine load, memory usage, error rate, network status), so passes (i3) of the monitor the
  • the decision maker unit analyzes the conflict between actual and target value displayed by the monitor with regard to the system availability and makes a decision based on suitable, adjustable criteria (e.g. through rule- or case-based programming) to resolve the conflict, the system status and thus to bring the availability of the overall system [5] back into the permissible range.
  • suitable, adjustable criteria e.g. through rule- or case-based programming
  • the decision maker then informs (i4) the so-called "decision enforcement” unit of the decision made to ensure the availability of the
  • the task of the decision enforcement unit is to implement this countermeasure against the applications concerned. For this purpose, a corresponding message is sent to the application identified as the cause of the limited availability.
  • a corresponding message is sent to the application identified as the cause of the limited availability.
  • non-AEM-compliant applications [4b] direct (7) as a system signal
  • AEM-compliant applications [4a] indirectly (6) as AEM API message.
  • this unit must be equipped with appropriate system priorities (eg belonging to the UNIX owner "root” and with a sufficiently high process / task priority, which in turn can be set using the "nice" UNIX system command.).
  • An AEM-compliant application [4a] requests (1,2) more real memory (RAM) via the AEM API [3a].
  • the AEM [3] concludes that this would impair the availability of the overall system to an unacceptable extent and rejects the request with an appropriate AEM API response (6).
  • the AEM-compliant application [4a] responds to this feedback with its event handler by using the slower virtual memory (in UNIX terminology "swap space") instead of the fast real memory (RAM) (after this has been previously approved by the AEM or may already have been proposed as part of the first AEM feedback (6) as an alternative).
  • a non-AEM-compliant application [4b] bypasses the AEM [3] by using a direct system call and threatens to jeopardize the availability of the overall system [5] with regard to the availability of network resources (network connectivity) because at the same time all other applications also require network services.
  • the AEM [3] recognizes this conflict in its resource management and the risk that the requirements may escalate due to a certain non-AEM-compliant application [4b] at the expense of the other applications [4a, b].
  • the problematic, non-AEM-compliant application [4b] is therefore temporarily stopped via the decision maker and decision enforcement units (eg via the UNIX system signal "STOP") until the permissible requests for network resources by the other applications result in such a maximum occupancy by an individual Allow application.
  • the availability are of the overall system in an unacceptable extent b eeintr thesisen would not, thus remains as a single agent before its final termination only a temporary stop as an alternative measure to the Ensure system availability.
  • AEM-compliant applications or AEM-compliant interactions of an application can react more optimally in such cases via their event handler (eg with reduced requirements for network resources).
  • the optimal two-way communication (1,2,6) between AEM [3] and AEM-compliant applications [4a] can be used to avoid radical measures such as the temporary stopping necessary for non-AEM-compliant applications [4b] or even the final termination . Stopping or scheduling to ensure system availability are only used for AEM-compliant applications [4a] if the treatment via the event handler does not produce a sufficient result.
  • the AEM subsystem [3c] can of course also monitor the above-mentioned interaction of AEM-compliant applications for the use of non-AEM-compliant methods and system calls, or monitor the aforementioned interaction of those applications that have been programmed, for example, with the aid of an older version of the AEM-API and now (ie if you look at applications programmed for comparison via a new API) are only partially AEM-compliant.
  • the invention has the following features / advantages for ensuring high availability in data processing systems:

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer And Data Communications (AREA)

Abstract

Pour des applications déterminées, par ex. pour des services de télécommunications, il faut que les systèmes informatiques requis à cet effet offrent, dans leur globalité, une disponibilité particulièrement grande, en d'autres termes dans la combinaison de la plate-forme informatique et des applications qui tournent dessus. A cet effet, il est prévu un élément de contrôle inséré entre les applications et la plate-forme informatique au niveau du logiciel.
PCT/EP1999/009055 1998-11-24 1999-11-23 Element de controle d'un systeme de traitement de donnees Ceased WO2000031639A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP99959313A EP1133729A1 (fr) 1998-11-24 1999-11-23 Element de controle d'un systeme de traitement de donnees

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
EP98122303 1998-11-24
EP98122303.5 1998-11-24

Publications (1)

Publication Number Publication Date
WO2000031639A1 true WO2000031639A1 (fr) 2000-06-02

Family

ID=8233027

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/EP1999/009055 Ceased WO2000031639A1 (fr) 1998-11-24 1999-11-23 Element de controle d'un systeme de traitement de donnees

Country Status (3)

Country Link
EP (1) EP1133729A1 (fr)
CN (1) CN1328667A (fr)
WO (1) WO2000031639A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010116441A1 (fr) * 2009-03-30 2010-10-14 富士通株式会社 Système d'alimentation électrique sans fil, dispositif de transmission de puissance sans fil et dispositif de réception de puissance sans fil

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101221615B (zh) * 2008-02-05 2011-08-17 北京飞天诚信科技有限公司 监测目标软件的方法和智能密钥装置

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996032675A1 (fr) * 1995-04-11 1996-10-17 Talati Kirit K Execution automatisee d'un comportement par un logiciel
EP0827077A1 (fr) * 1996-07-01 1998-03-04 Sun Microsystems, Inc. Système, méthode et article de fabrication orienté objet pour un processus client-serveur de communication d'erreurs

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO1996032675A1 (fr) * 1995-04-11 1996-10-17 Talati Kirit K Execution automatisee d'un comportement par un logiciel
EP0827077A1 (fr) * 1996-07-01 1998-03-04 Sun Microsystems, Inc. Système, méthode et article de fabrication orienté objet pour un processus client-serveur de communication d'erreurs

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2010116441A1 (fr) * 2009-03-30 2010-10-14 富士通株式会社 Système d'alimentation électrique sans fil, dispositif de transmission de puissance sans fil et dispositif de réception de puissance sans fil

Also Published As

Publication number Publication date
EP1133729A1 (fr) 2001-09-19
CN1328667A (zh) 2001-12-26

Similar Documents

Publication Publication Date Title
DE69614383T2 (de) Kontinuierlich verfügbarer datenbankserver mit mehreren knotengruppen mit sich minimal überschneidenden sätzen von datenbankteilkopien
DE69032673T2 (de) Verfahren und Vorrichtung zur dynamischen Verwaltung der Eingabe-/Ausgabeverbindungsmöglichkteiten
DE60220418T2 (de) Verfahren und Anbieter zur Systemsynchronisation
DE69623229T2 (de) Bindungsverfahren in einer Transaktion in einer verteilten Datenbank
DE60207251T2 (de) Verfahren zur sicherstellung des betriebs eines gruppierten nachrichtenvergebenden servers während knotenausfällen und netzwerkzuteilungen
DE60318468T2 (de) Verfahren zur lösung von entscheidungslosigkeiten in einem cluster-rechnersystem
DE10321454B4 (de) System und Verfahren zur Leistungsverwaltung in einem Computersystem mit mehreren Stromversorgungsnetzen
DE2740056A1 (de) Mulitprozessor-rechnersystem
DE2908316A1 (de) Multikonfigurierbares modulares verarbeitungssystem, das mit einem vorverarbeitungssystem integriert ist
EP0959588A2 (fr) Elément de réseau avec le dispositif de commande et la méthode de contrÔle
DE112011103443T5 (de) Intelligente Schnittstelle für ein dezentrales Steuerungssystem
DE102016219854A1 (de) Computersystem und Verfahren zum dynamischen Anpassen eines software-definierten Netzwerks
DE4429969A1 (de) Verfahren für einen Programmpaketeaustausch in einem Mehrrechnersystem und Rechner dafür
DE102009004726A1 (de) Systeme und Verfahren zum Verfolgen von Befehlszeigern und Datenzugriffen
DE602004002241T2 (de) Schutz eines auf ausführungwartenden Programms in einem Speicher für einen Mikroprozessor
DE102007005207A1 (de) Software-Duplikation
EP0433350A1 (fr) Programme d'exploitation d'un ordinateur
WO2000031639A1 (fr) Element de controle d'un systeme de traitement de donnees
DE69615460T2 (de) Verfahren und Vorrichtung zur Zuteilung von Transaktionsidentifikatoren
EP1536328B1 (fr) Système de traitement de données avec la gestion automatisable et procédé pour la gestion automatisée d'un système de traitement de données
EP0850545A1 (fr) Systeme environnemental operationnel pour des applications de services d'un reseau de telecommunication
DE69800095T2 (de) Schnelles Semaphoreregister mit einer sicheren Arbeitsweise ohne spezifisches Busprotokoll
DE1966991A1 (de) Ausfallgesicherte datenverarbeitungsanlage
LU101163B1 (de) Verfahren und Vorrichtungen für eine Lastzuweisung und Überwachung für eine zuzuweisende versorgungssicherheitskritische Ressource in einem Netzwerk
EP1195678A2 (fr) Méthode d'opération d'un système de traitement de données avec une unité de traitement redondante

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 99813681.6

Country of ref document: CN

AK Designated states

Kind code of ref document: A1

Designated state(s): BR CN ID US

AL Designated countries for regional patents

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LU MC NL PT SE

121 Ep: the epo has been informed by wipo that ep was designated in this application
DFPE Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101)
WWE Wipo information: entry into national phase

Ref document number: 1999959313

Country of ref document: EP

WWE Wipo information: entry into national phase

Ref document number: 09856629

Country of ref document: US

WWP Wipo information: published in national office

Ref document number: 1999959313

Country of ref document: EP

WWW Wipo information: withdrawn in national office

Ref document number: 1999959313

Country of ref document: EP