United States Patent
`Matsumoto
`
`[54] COMPUTER OPERATION MANAGEMENT
`SYSTEM FOR A COMPUTER OPL'RATING
`SYSTEM CAPABLE Olr SIMULTANLOUSLY
`FXFCUTINC PLURAI, APPI,ICATION
`PROC HAMS
`
`[75[
`
`Inventor Hgjime Matsumnto, Iiami, Japan
`
`[73] Assigncc: Mitsubishi Denki Kabushiki Kaisha,
`Tokyo, Japan
`
`*
`
`[
`
`] Notice:
`
`This patent issued on a continued pros-
`ecution application file under 37 CFR
`1.33[d), and is subject to thc twenty year
`term provision«of 35 U.S C.
`patent
`154[a)(2)
`
`[Zl] Appl. No: 539,007
`Oct. 4, 1995
`I iled:
`Foreign Application Priority Data
`
`[22]
`
`[30]
`
`May 31, 1995
`
`[JP[
`
`Japan .................................... 7 133920
`
`Int. Cl.'............... ('06F 9/46
`
`[51]
`[52[ U.S. CL ...................... 395/672; 395,'674, 395)677
`[58] Field of Search .................................. 395/650, 672,
`395 674, 677, 364/~81.7, 281.8, 281 4,
`281, 281.6, 281.'3, 285.3, 286 4, 265
`
`[56]
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`4B)8,173
`4,493.020
`4.845,644
`
`3/1982 I'reedman ct at
`1,)198«K&m et al...........
`7)1989 Aa&hiss a& al.
`
`364/21st
`364,'200
`364/521
`
`IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII
`US005835765A
`[i i] Patent Number:
`l)ate of Patent:
`
`5t S351765
`aNov. 10, 1998
`
`[45]
`
`«,210,872
`5,421,0) 1
`«,-137,032
`«,440,726
`«,-142,791
`«,46E354
`«,-185,62G
`5,495 &&0&i
`
`«,'1993
`5 1995
`7,'1995
`8 1995
`8,'1995
`tt 1995
`1,'1996
`2)09&r&6
`
`Ferguson et al.
`('emit)one et at.
`«Volt et al...........
`1'uchs et at
`«Vrahetz et al.
`Hiic«awa et at
`Lawlo& et al.
`Borden et al.......
`
`..... 395,)GSO
`39:i/6:it)
`..... 395,)GSO
`. 39«/82 )8
`..... 395,)GSO
`39:i/6:it)
`..... 395,)GSO
`..... 395/600
`
`OTHER PUBLICATIONS
`I.awrence .I Kenah et al., "VAX VMS Internals and Data
`Structures'*, Digital Pre««,
`3—12, 183 — 188,
`1984, pp
`443 —517 and 704—704.
`Gcndrcav ct al, Cooperative Self Scheduling In Mcssagc
`Passing Parallel Systems, IEEE Conf. Proceed. 1989, pp.
`496-499.
`Genrlreav, Process Management Issues In Message—Based
`Mult&processor«, ILLL''onf. Proceed 1988, pp 103—107
`Stumm, I'be De«ign and Implementation of a Decentralized
`Scheduling Facility For A Workstation Cluster, Comp.
`Workstatinn«Conf. 1988 pp. 12—22
`Pris&an Exam/ocr~un&en U Toplu
`Ar/orna)i Agenr, or Fir&n~cuermott, Will a Lmery
`ABSTRACT
`[57]
`
`A computer operatinn management system comprises a
`computer having a central processing unit, a main memory
`and, an auxiliary memory, an operating sy«tcm for operating
`the computer so as for thc computer to cxccutc plural
`applicauon, pro rams simultaneously and a process man-
`ager for monitoring an execuuon status of plural application
`programs to control the number of application programs
`simultanenusly executed and their execution priorities
`
`11 Claims, 14 Drawing Sheets
`
`11 Computer operation management system
`
`21
`
`12
`
`Google Exhibit 1053
`Google v. VirtaMove
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 1 of 14
`
`5,S35,765
`
`Fig. f
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 2 of 14
`
`5,S35,765
`
`«D
`Ui
`
`'PJ
`
`N (
`
`DEQC
`
`D
`
`E 0 (
`
`DC
`
`L
`
`O LQ
`
`I
`
`I
`
`I
`
`I I I I I
`
`I
`
`) OE00
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 3 of 14
`
`5,S35,765
`
`Computer operation management system
`
`Fig.3
`
`tion
`
`Other computers
`
`34
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 4 of 14
`
`5,S35,765
`
`Execution controiier
`
`START
`
`S41
`
`S42
`
`S43
`
`S44
`
`S45
`
`S46
`
`+END
`
`ter
`ime
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 5 of 14
`
`5,S35,765
`
`Ex. state monitor
`
`SSO START
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 6 of 14
`
`5,S35,765
`
`Fig.6
`
`Execution priority controller
`
`START
`
`S61
`
`S62
`
`S63
`
`S64
`
`S65
`
`+END
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 7 of 14
`
`5,S35,765
`
`Fig. 7
`
`Ex. priority controller
`
`S71
`
`S72
`
`S73
`
`END
`
`Fig. 8
`
`Error recovery processor
`
`START
`
`S81
`
`S82
`
`S83
`
`S84
`
`

`

`U.S. Patent
`U.S. Patent
`
`Nov. 10, 1998
`Nov. 10, 1998
`
`Sheet 8 of 14
`Sheet 8 of 14
`
`5,S35,765
`5,835,765
`
`/Bulsse00l¢4@dnoiryew|ou£00|
`
`
`ainpecoidAle@Aod9sJOEJOUOIUYEG
`
`weiboidpejyejel
`
`weiboid
`
`joBulssao0id
`Ds}eE|ay
`
`
`
`ainsealy
`JOOJO}
`Kso6e720'Bold
`
`Ou
`
`Ou
`
`AlQA0991
`
`ssa00id
`
`A\q!910}
`
`Bulpus
`
`6bly
`
`
`
`
`
`
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 9 of 14
`
`5,S35,765
`
`Fig. 1 0
`
`Ending processor
`
`START
`
`S101
`
`S102
`
`S103
`
`S104
`
`S105
`
`S106
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 10 of 14
`
`5,S35,765
`
`Computer resource manager
`
`START
`
`

`

`U.S. Patent
`
`Nov. 10,1998
`
`Sheet 11 of14
`
`5,S35,765
`
`Fig. 12
`
`Network manager
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 12 of 14
`
`5,S35,765
`
`Error communication means
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 13 of 14
`
`5,S35,765
`
`Error communication means
`
`Checking
`periodically
`
`S1
`
`S1
`
`S1
`
`S1
`
`S1
`
`

`

`U.S. Patent
`
`Nov. 10, 1998
`
`Sheet 14 of 14
`
`5,S35,765
`
`Computer operation management system
`
`Fig. f5
`
`07
`
`08
`
`ion
`
`Other computers
`
`204
`
`

`

`COMPUTER OPERATION MANAGEMENT
`SYSTEM FOR A COMPUTER OPERATING
`SYSTEM CAPABLE OF SIMULTANEOUSLY
`EXECUTING PLURAL APPLICATION
`IROGRAMS
`
`I
`
`lc
`
`zn
`
`BACKGROUND OF THE INVENTION
`I'ield of the invention
`Thc prcscnt invention rclaks to a system enabling hi h
`reliability cxccution and high spccd processing of plural
`application programs execunng on a single computer or on
`plural computers over a network, and relates particularly to
`a stable operation management system lor computers.
`2. Description of thc prior art
`Conventional computer systems comprise harclware such
`as the memory, disk dnves, central processing unit (CPU),
`and network devices, operanng system software enabhn ~
`the hard&vare to be used effectively, ancl programs recording
`the prncedures to be executed by the computer to accom-
`plish specific tasks. Thc person using thc computer typically
`executes application programs written in a programmin
`language by the pro rammer by instrucnng the operann ~
`system software to execute, or "run*', the desired application
`program by some means. The operating system then
`executes the one requested program, or simultaneously
`executes plural programs, by effectively using the memory
`(primary storage means), disk drive (complcmcntary storage
`means), CPU, and network clcviccs availablc to thc com-
`puter. Some of the de facto standard, operann ~ systems in
`use, today include MVS used on mainframe computers, VAX
`used on minicomputers, UNIX used on workstations, arnl
`MS-DOS and Windows used on personal computers
`Of these common operating systems, UNIX is capable of
`simultaneously execut&ng plural application programs.
`UNIX was originally clcvclopcd by AT&T of thc Unitccl
`States, and is widely used on primarily worl stations today.
`Dug&nally clesigned for soltware development, computer-
`aided design (CAD), and other technical applications run by
`engineers, UNIX olfers the lollowing leatures in uimpanson
`with the other operating systems mentioned above
`(I) UNIX can simultaneously execute plural applicat&on
`programs in parallel
`(3) Onc program cxccuting under UNIX can cause
`another program to bc run under UNIX, and thc number of
`simultaneously cxccuting programs can incrcasc without
`being limited by the operanng system.
`(3) Simultaneously execunng programs are harnffecl
`equally and execute with the same pnority unless one or
`to execute with precedence
`more pro rams is specifiecl
`(priority) over another program(s). 'I'he priority of execut&ng
`programs can also bc changccl dynamically.
`(4) Because thc virtual storage area has a finite size,
`attempts to cxccutc a program cxcccding this limit will ss
`cause the program to end immediately arnl an error to be
`generated (an "abnormal eral").
`The problem with UNIX is that the number of applicanon
`programs simultaneously executable on a single computer
`cannot be limited As a result, when the number nf simul-
`taneously cxccuting programs cxcccds thc projcctcd limit,
`the capacity of thc main storage dcvicc (memory) becomes
`insufficient. W'hen memory becomes insullicient, additional
`virtual storage area is reserved on the complementary stor-
`age device (dial dnve), and the data in memory is wntten to es
`disk When the virtual storage area (virtual memory) is used
`for prngram execution, however, the executing speed of the
`
`zs
`
`program drops by several ten to several hundred percent.
`With some apphcanon programs it is merely inconvenient
`when thc cxccution speed drops, but with syskms requiring
`real-time performance, rcduccd cxccution speed can bccomc
`a source nf trouble.
`When both hardware memory and virtual memory are
`insufficien, UNIX forcibly tcrminatcs onc of thc cxccuting
`programs. Bccausc it is not possible to specify thc order
`(prccedcncc) in which programs arc forcefully krminatcd,
`programs for which reliability is csscntial and pro rams for
`which reliability is non-essennal will be lorcibly termmated
`with the same probability. When any program is forcibly
`terminated, system functionality is also lost
`UNIX also does not have any means of centrally manag-
`ing what changes have occurred in any program, when those
`changes occurred, and what error messages were generated
`when a program starts up, terminates normally, or terminates
`abnormally. As a result, when a system error or problem
`clcvclops, there is no way to dctcrminc v hat program was
`the source ol the problem. While some error messages are
`the messages are not un&lied by machine (CPU),
`output,
`output device, or output method, and troubleshoonng is
`therefore difficult and time-consuming
`UNIX also handles each of the plural simultaneously
`executing programs equally, arnl is not able to control the
`execution time (CPU time) allocated to each prngram As a
`result, programs that make unusually heavy clemands on the
`CPU or memory degrade thc overall rcsponsc time and
`throughput of the system.
`In large scale systems linking UNIX based computers via
`a common network, the problems dcscribcd above can occur
`frcqucntly, and tracing thc source of thc problem is cvcn
`more comphcated. There is aLso no means of immediately
`notifying the operator that one of aforemennoned problems
`has occurred, and recovery efforts are therefore delayed.
`SUMMARY OF THE INVENTION
`Therefore, an obt act of the present invennon is to provnle
`a computer operation management system that operates on
`an operating system, such as UNIX, having the aforemen-
`tioned problems.
`I'o achieve the aforementioned object, a cnmputer opera-
`tion management system accorcling to the present invention
`cnablcs stable and high reliability cxccuhon of individual
`application programs by limiting thc number of simulta-
`neously cxccutablc apphcation programs currently
`executing, arnl momtoring apphcanon program execution,
`computer resources, and the network in a computer operat-
`ing system capable of simultaneously executing plural appli-
`cations programs
`A computer operation management system according to
`first aN&ect of the inventinn resides between the operating
`system and the simultaneously executing application pn&-
`grams has an intcrfacc to both thc operating system and thc
`individual application programs, ancl appears to thc operat-
`ing system to be a smgle apphcation program. This com-
`puler operation management system composes a program
`clelinition file predelinecl by the user to record the maximum
`number of simultaneously executable prngrams for each
`prngram category, the execution prionty of the program, the
`error recovery proccdurc, programs that cxccutc by com-
`munication with other programs on thc network, and infor-
`the post-process to be executed after the
`mation about
`program terminates,
`a queued program record tile for
`recording the information of queued programs wainng to be
`hie fnr recording log, clata when a
`executed; and a lo
`prngram starts or ends, and when an errnr occurs.
`
`

`

`A computer operation management system according to
`second aspect of the invention manages and controls the
`execution status of application programs by executing apph-
`cation programs to be manage&1 in a manner enabling direct
`communication with or direct management by onc operation
`manigcmcnt program for managing thc operation of them.
`A computer operation management system according to
`third aspect of the invent&on achieves the application pro-
`gram control method by means of a program definition file
`for rccorrling the maximum number of simuhancously
`executing programs for each program category or computer
`as delined bit the user, an execution controller for elfectively
`using computer resources and achieving stable apphcation
`program operation during simultaneous execution of plural
`programs hy limiting for the operating, system at applicat&on
`program startup the number of executing programs by
`program category or computer based on thc maximum
`number of simultaneously cxccutablc pro rams read from
`the pro ram delinnion lile, a queued program record tile for
`temporanly storin ~ the data of programs which cannot be
`currently simultaneously executed to enable the queued
`program to he started after a prerleterm inc&I period; and a log
`file for recordmg after prngram startup a startup log con-
`taining thc name of thc started program, thc time anti date,
`and thc computer name.
`A computer operation management system according to
`fourth aspect of the inventinn achieves the applicat&on
`program status monitonng method by means of an execution
`status monitoring means comprising: a means for regularly
`sending a status rcqucst mcssagc to and rccciving a reply
`from thc application program, a means for confirnung that
`an application program is executing by receiving the reply
`from the applicauon program and conlirming the content of
`the received reply; and a means fnr rletermining a program
`error when there is no response from the application pro-
`gram to the execution status request message within a
`predctcrmincd penod, and then notifyin thc error recovery
`processor executing thc prcdctcrmincd error
`rccovcry
`process, aml the error communicanon means for commum-
`cating when an error occurs.
`A computer operation management system according to
`fifth aspi,ct of thc invention achicvcs thc error recovery
`method cxccutcd when an application program error occurs
`by the program deliniuon lile recording the error responses
`method predelined by the user for execution when a program
`an&1 a recovery process execuun ~
`terminates abnormally,
`means for executing error recovery when an error occurs
`based on the error response method recorded to the program
`definition fil.
`A computer operation management system according to
`sixth aspect of the invention achieves the method for com-
`municating errors occurring on the computer to the operator
`by the program dehnition file recnrrling the error commu-
`nication method dehned by the user; and an error commu-
`nication means comprising: a means for collcctin thc lo
`files of applications software under thc mana cmcnt of the
`operahon management system on all computers connected
`to the network, a means for processing an&1 displaying data
`for the operator in an easy-to-unrlerstand format; and a
`means for automatically indicating an error by means of
`issuing an alarm, electronic mail, a consnle, printer, and/or
`a fik.
`A computer operation management system according to
`seventh aspect of the invenuon achieves the method for es
`simultaneously executing plural programs under the oper-
`ating system hy receiving new program startup requests m
`
`35
`
`&0
`
`40
`
`the execuuon status mon&loon ~ step alter the execution
`control and process startup step is completed; anti startiog
`inrlependently and parallel ivith previously executed pn&-
`grams that are subsequently started
`A computer operation management system according to
`the invenuon achieves the methotl for
`eighth aspect of.
`controlling Ihe, execution pnonty of the applicauon pro-
`grams by: the program definition file recording, the execution
`pnonty sequence of each program category as defined by the
`user; and an execution prionty scqucncc controller compris-
`ing a means for dynamically controlling thc priority
`sequence of both executing and queued programs based on
`the prionty sequence read from the program delinition lile
`by resequencing the queued programs according to the
`pnonty sequence recorded in the queued program record
`file, dynamically changing the execution priority nf execut-
`ing pro rams and thc cxccution priority of qucucd programs,
`and informing the operating system of thc changed priority
`sequence.
`A computer operation management system accnrrling to
`n&nth aspect of the invention achieves the methocl
`for
`stnpping an application program executing nn a computer
`by: the program definition fik recording thc definitio of a
`post-process define&1 by the user to bc subscqucntly cxccutcd
`according to the en&ling status, and an ending processor
`comprising a means for recogmzing program termination
`based on an ending message from the operating system, a
`the ending status when a program
`means for identifyin
`terminates, a post-process executing means for executing the
`post-process according to thc post-process definition, and a
`log fil recorrling thc cnd log when a program ends, said cnd
`log containing the program name, computer name, anil ume
`and &late of program terminauon.
`A computer operation management system accnrrling to
`tenth aspect of thc invention achicvcs thc method for man-
`aging thc resources of thc application program cxccuting on
`the computer by: the program delinnion tile recording for
`each execunng application program the maximum primary
`sti&rage device (memory) capacity, complementary siorage
`device (disk drive) capacity, and cumulative processor time
`limit usable by each program; and a computer resnurces
`mana cr comprising a means for dctccting when a program
`is executed excccding thc rlcfincd primary storage and
`compkmentary storage device capacity, or the cumulauve
`pri&cessor time limit, and a means having a lunction for
`iletermining an error when the computer resources are used
`exceeding the defined limits, and notifying, the error recov-
`ery prncessor and the error commun&cat&on means.
`A computer operation managcmcnt system accordin to
`ekventh aspect of the mvenuon achieves the method for
`managing Ihe operahon of applications software executed by
`means of communications between plural computers con-
`nected to a common nenvork is achieved by the program
`dehnition file recording information defined by the user for
`programs that cxccute by association and communication
`over a network; and a computer network manager compris-
`ing a means for communicaung between the operation
`management systems installed on diferent computers on the
`netwnrk, a means for managing the operatinn of programs
`that execute by associatinn and communicatinn over a
`netwnrk according to the information for said programs
`rlcfincd in the program definitio file, ancl a means for
`recognizing combinations of programs that cxccutc by asso-
`ciation and communication over a network.
`A computer operation management system accnrrling to
`twelfth aspect of the invention executes the execution
`
`

`

`&5
`
`controller, execuuon status moniloring means, computer
`resources manager, aml ending processor as a conunuous
`process within the main program, and cxccutcs the execu-
`tion pnorily sequence controller, error recovery processor,
`error communication means, aml nelworl manager
`on-demaml as imlependenl programs by the mam program.
`A computer operation management system according to
`the invenhon executes Ihe execution
`thirteenth aspect ol.
`controller, execution status monitoring means, and ending
`processor as a continuous process within the main program,
`and executes the execution priority sequence controller,
`error recovery processor, error communication means, com-
`puter resources manager, and network mana cr on-dcmantl
`as inilepemlent programs by Ihe main program.
`Opera&&on:
`A computer operation management system according to
`the invention limits the number application programs
`executing on an operating system enabling the simultaneous
`execution of plural apphcation programs, and monitors
`application program cxccution, computer rcsourccs, and the zo
`network to enable stable, high reliability operauon of inth-
`vidual applicahon programs.
`A computer operation management system according to
`the hrst aspect of the invention resirles between the simul-
`taneously executing application programs anil the operating
`system, which is charactcrizcd by being able to simulta-
`neously cxccute plural application programs, being able to
`generate from one executing program a new relate&i program
`that can be communicated with or managed directly, han-
`dling all executing programs equally unless the operating
`system is othe&w&se specifically notifierl during program
`startup, and dynamically changing the priority sequence of
`executing programs, having a virtual storage arcs of finite
`size, and immctliatcly abnormally ending program cxccution
`when the program to be executed exceeds the lrmits of the zs
`virtual storage area. The computer operauon management
`system further has an interface to both Ihe operatm ~ system
`and the individual application programs, and appears to the
`operatin'ystem to be a single application program The
`program definition tile is prcdcfincd by thc user to rccortl the ao
`maximum number of simultaneously cxccutablc programs,
`the execuuon pwonty of Ihe program,
`the error recovery
`procedure, programs that execute by commun&canon with
`other programs on lhe network, and informauon about the
`post-process to he executed after the program terminates for
`each program category A queued program record file
`records thc information of qucucd programs ivaiting to be
`cxecutctl; and a log fil records log data when a program
`starts or ends, aml when an error occurs.
`A computer operation management system according to
`the secoml aspect of the invenuon manages and controls the
`execution status of application programs by executing on
`one operating system one operation management program
`for managing thc operation of plural application programs in
`a manner cnabhng tlircct communication with or direct
`mani ament of lhe applicauon programs to be mana ed.
`A computer operauon management system accordrng to,
`the thiril aspet:I of the invention records to the program
`definitions file during application program startup the maxi-
`mum number of simultaneously executing programs for
`each program category or computer as dcfincd by thc user.
`During simultaneous cxccution of plural application
`programs, the execution controller limits for lhe opera&in ~
`system the number of executing programs by program
`category or computer based on the maximum number of as
`simultaneously executable programs read from the program
`definition file (program defmitions recortling means) As a
`
`O
`
`s
`
`result, computer resources can be elfecuvely uulized, and
`stable operation of the individual apphcation programs is
`achieved. When a program cannot currently be simulta-
`the execution controller temporarily
`neously executed,
`stores thc data of that program to a qucucrl program record
`file to cnalik thc queued program to bc cxccutcd after a
`pre&le&arm&natl per&od. After a program is started, the execu-
`uon controller records lhe name of the slarled pro ram, the
`time and date, and the computer name to a log tile
`In a computer operation management system according to
`the fourth aspect of the invention,
`the execution status
`monitoring means regularly sends an cxccution status
`rcqucst mcssagc to and rcceivcs a reply from thc application
`pri&gram, aml conlirms the content of the received reply to
`&le&ermine whether the applicauon program is executing.
`When there is no response from the application program to
`the execution status request message, an error is determined
`to have occurred, and the error recovery processor executing
`the prcdctermincd error rccovcry process, and thc error
`communication means for communicating when an error
`occurs, are nolilied.
`In a computer operation mana ament syslem accordmg to
`the fifth aspect of the invention, the program definition file
`recorris the error response method predefined by the user for
`execution when a program terminates abnormally, and a
`recovery process cxccuting means cxccutcs error rccovcry
`when an error occmrs based on thc error rcsponsc method
`rerul from the program deiiniuon tile.
`In a computer operation mana ament syslem accordmg to
`the sixth aspect of the invention, the program detinition file
`recorris the error com mun icat ion method defined by the user
`The error communication means collects the log files of
`applications software under thc managcmcnt of the opera-
`tion management system on all computers conncctcd to thc
`network. Based on the method read lrom the program
`ileiinition lile, data is processed and displayed in an easy-
`to-umlerslaml format for Ihe operator, the operator is auto-
`matically notified by means of issuing an alarm, electronic
`mail, a console, printer, and/or a file
`In a computer operation man age ment system according to
`the seventh aspect of thc invention, ncw program startup
`requests are receive&i in the execution status monitoring slap
`the execution control and process star&up step
`after
`completed, aml the last-executed programs are started inde-
`pendently and parallel
`ivith any programs previously
`executed to achieve simultaneous execution of plural appli-
`cation programs.
`In a computer operation man age ment system according to
`eighth aspect of lhe invention, Ihe program delinilion lile
`records Ihe execuuon priority of each program category as
`ileiined by lhe user. Based on the prionty sequence read
`from the program detinition tile,
`the execution priority
`sequence controller dynamically controls the priority
`scqucncc of both executing and qucucd programs by rcsc-
`qucncing the qucucd programs according to the priority
`sequence recorded in the queued program record lile,
`the execuuon pnonty of executing
`ilynamically changin ~
`pri&grams and lhe execution pnonty of queued pro rams,
`anti notifying the operating system of the changed priority
`sequence
`In a computer operation man age ment system according to
`the ninth aspect of thc invention, thc program dchnition file
`records Ihe delimtion of the post-process deiined by the user
`to be subsequently executed according to the ending status.
`The ending processor recogmzes that a program has termi-
`nated based on an ending message from the operating
`identifies the ending status of the program, and
`system,
`
`

`

`executes the post-process according to the definition re&xi
`from the program definition file. When a program
`terminates, an end log cents&ning the program name, com-
`puter name, anil time and elate of prngram termination is
`recorded to thc log file.
`In a computer operation management system according to
`the tenth aspect of: Ihe invention, the program drain&&ion file
`records for each execuung application program the maxi-
`mum primary storage device (memory) capacity, comple-
`mentary storage rlevice (disk drive) capacity, and cumulative
`processor time limit usable by each prngram. 'I'he computer
`resources manager dctccts when a pro ram is cxccutctl
`cxcccding thc limits dcfincd in thc program definition file,
`determines an error when the computer resources are used
`exceeding the defined limits, and notifies Ihe error recovery is
`processor and the error communicatinn means when an error
`is determined
`In a computer operation management system according to
`the ckventh aspect of thc invention, the program definition
`file records information for programs that cxccutc by asso-
`ciauon and communicauon over a nelworl
`. The computer
`network manager commun&ca&ca between the operanon
`management systems installed on different computers on the
`netivork, manages the operatinn of applications software
`using the network accorrhng to the information defined m
`the pro ram dctinition tile, and rccognizcs combinauons of
`programs that cxccutc by association and communication
`over a nelworl .
`In a computer operation managemenl system according lo
`the twelfth aspect of the invention, the execution controller,
`execution status mon&toring means, computer resources
`manager, and ending processor are executed as a continuous
`process within thc main program, and thc cxccution priority
`scqucncc controller, i'ffof fccovcfv processor, error commu-
`nicauon means, aml network manager are called on-demaml zs
`and executed by the main program as imlependenl programs.
`In a computer operation managemenl system according lo
`the thirteenth aspect of the invention,
`the execution
`controller, execution status monitoring means, and ending
`processor arc cxccutccl as a continuous process within the ao
`main program, and thc execution priority sequcncc
`controller, error recovery processor, error commun&canon
`means, computer resources manager, and network manager
`are called on-demand and executed by the main program as
`independent programs.
`
`as
`
`zo
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`The present invention will become more fully understood
`from the detailerl descript&on given below anil the accom-
`panying diagrams whcrcin:
`FIG. I is a conceptual rliagram of thc relationship between
`the operating system, application programs, and thc com-
`puler operation management system of the present inven-
`tion,
`FIG. 2 is a block diagram of Ihe computer operauon
`management system accorrling to the present invention;
`I'IG 3 is a block d&agram anil floiv chart of the vanous
`controllers in one embodiment of a computer operation
`manigcmcnt system accordin to thc prcscnt invention;
`FIG. 4 is a floiv chart of thc operation of thc cxccution
`controller in onc cmbodimcnt of a computer operation
`man i ament system according lo lhe present inventron,
`FIG. 5 rs a flow chart of the operauon of the execuuon
`status momtoring means in one embodiment of: a computer ss
`operation management system according to the present
`invention;
`
`ss
`
`FIG. 6 is a flow chart of: the operauon of: Ihe execution
`priority sequence controller ivhen program startup is
`requested in the computer operat&on management system
`according to the present invention,
`FIG. 7 is a flow chart of: the operauon of: Ihe execution
`pnonlv sequence controller when Ihe pnonly sequence of
`executing programs is dynamically change&1 in the computer
`operation management system according,
`to the present
`invention;
`FIG. 8 is a flow chart of thc operation of thc error rccovcry
`processor in ihe computer operauon management system
`according to lhe present invention,
`I lG 9 is a table of one example of the procedural
`tlcfinitions for error rccovcry processing in thc computer
`operation management system according to thc prcscnt
`invention;
`I lG 10 is a flow chart of the operation of the eedli&ig
`prncessor in the computer operat&on management system
`accordin to the present invention;
`FIG. 11 is a flow chart of: Ihe operation of Ihe computer
`resources manager in Ihe computer operation management
`system according the present mvention;
`FIG. 12 is a fiovv chart of thc operation of the network
`mana cr in thc computer operation managcmcnt system
`according to Ihe prevent invention,
`I lG 13 is a flmv chart of the operatinn of the em&r
`communication means achieving the errnr communication
`function of the computer operation management system
`accordin to the present invention
`FIG. 14 is a flow chart of: the operauon of: the error
`communication means achieving the error communication
`function and the error monitoring functinn of the computer
`operation management system according,
`to the present
`invention; and
`FIG. IS is a flow chart and block diagram of Ihe vanous
`conlrollers in a secontl embodiment of a computer operation
`management system according to the present invention.
`
`DESCRIPTION OF PREFERRED
`EMBODIMENTS
`The preferred embodiments of: a computer operation
`management system according Io Ihe present invenuon are
`described hereinbelow with reference to the accompanyi&ig
`figures
`Embodiment 1:
`FIG. I is a conceptual diagram of thc software operation
`management system in a computer system according Io the
`invention. As shown in FIG. 1, The computer
`present
`operation management system 1 of. Ihe presenl invention
`resides between the operating system 2 and the application
`prngrams 3, functioning as a single applicatinn program
`from the perspcctivc of the operating system 2, but with an
`intcrfacc to the application programs 3 that makes thc
`computer operanon management system I appear lo be par&
`of the operating system 2 from the perspective of the
`applicauon programs 3. The funcuon of Ihe computer opera-
`tion management system I is to monitor and control execu-
`tion of the application programs 3
`This operating system 2 has thc following characteristics.
`(I) Thc aliility to cxccutc plural apphcalion programs in
`parallel
`(2) The ability to generate a new program (child program)
`from one execuun ~ program (parent program); the number
`of simultaneously executing programs increases without
`heing limited by the operating system
`
`

`

`&0
`
`zo
`
`so
`
`is
`
`(3) Simultaneously execuung programs are executed aml
`handled with the same p