A hybrid solution of fork/join synchronization in parallel queues
A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K/spl ges/2 . The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronizati...
Saved in:
Published in | IEEE transactions on parallel and distributed systems Vol. 12; no. 8; pp. 829 - 845 |
---|---|
Main Author | |
Format | Journal Article |
Language | English |
Published |
New York
IEEE
01.08.2001
The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K/spl ges/2 . The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronization in each branch. Jobs arrive with mean rate /spl lambda/ and a general arrival distribution. Upon arrival, a job forks into K tasks. Task k, k=1, 2, ..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A job leaves the HFJ system as soon as all its tasks complete their service. In other words, tasks corresponding to the same job are joined before departing the HFJ system. We obtain a general and simple hybrid solution which combines analysis and simulation for the mean response time that we denote by T/sub K/. We obtain a very simple (as a function of T/sub 1/ and T/sub 2/ only) and general upper bound expression for T/sub K/ and we get an exact relationship between the cases for K=2 and 3. We evaluate our results by simulating 2, 3, ..., 99, and 100 queues for p=0.1, 0.2, ...0.8, and 0.9, each for four different HFJ cases, where /spl rho/=/spl lambda///spl mu/ and /spl mu/ is the average task service rate for a server. The maximum absolute offset for our hybrid solutions from all the simulations is less than 0.33 percent (1/300), which is a reasonable error ratio for simulation. The maximum offset for our upper bounds over all the simulations is 21 percent. |
---|---|
AbstractList | A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K greater than or equal to 2. The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronization in each branch. Jobs arrive with mean rate lambda and a general arrival distribution. Upon arrival, a job forks into K tasks. Task k, k = 1, 2,..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A job leaves the HFJ system as soon as all its tasks complete their service. In other words, tasks corresponding to the same job are joined before departing the HFJ system. We obtain a general and simple hybrid solution which combines analysis and simulation for the mean response time that we denote by T sub(K). We obtain a very simple (as a function of T sub(1) and T sub(2) only) and general upper bound expression for T sub(K) and we get an exact relationship between the cases for K = 2 and 3. We evaluate our results by simulating 2, 3,..., 99, and 100 queues for rho = 0.1, 0.2,..., 0.8, and 0.9, each for four different HFJ cases, where rho = lambda / mu and mu is the average task service rate for a server. The maximum absolute offset for our hybrid solutions from all the simulations is less than 0.33 percent (1/300), which is a reasonable error ratio for simulation. The maximum offset for our upper bounds over all the simulations is 21 percent. Also, we compare our results with three recent papers. Task k, k=1, 2, ..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K 2 . The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronization in each branch. Jobs arrive with mean rate lambda and a general arrival distribution. Upon arrival, a job forks into K tasks. Task k, k=1, 2, ..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A job leaves the HFJ system as soon as all its tasks complete their service. In other words, tasks corresponding to the same job are joined before departing the HFJ system. We obtain a general and simple hybrid solution which combines analysis and simulation for the mean response time that we denote by T(K). We obtain a very simple (as a function of T(1) and T(2) only) and general upper bound expression for T(K) and we get an exact relationship between the cases for K=2 and 3. We evaluate our results by simulating 2, 3, ..., 99, and 100 queues for p=0.1, 0.2, ...0.8, and 0.9, each for four different HFJ cases, where rho=lambda/mu and mu is the average task service rate for a server. The maximum absolute offset for our hybrid solutions from all the simulations is less than 0.33 percent (1/300), which is a reasonable error ratio for simulation. The maximum offset for our upper bounds over all the simulations is 21 percent A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K greater than or equal to 2 . The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronization in each branch. Jobs arrive with mean rate lambda and a general arrival distribution. Upon arrival, a job forks into K tasks. Task k, k=1, 2, ..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A job leaves the HFJ system as soon as all its tasks complete their service. In other words, tasks corresponding to the same job are joined before departing the HFJ system. We obtain a general and simple hybrid solution which combines analysis and simulation for the mean response time that we denote by T sub(K). We obtain a very simple (as a function of T sub(1) and T sub(2) only) and general upper bound expression for T sub(K) and we get an exact relationship between the cases for K=2 and 3. We evaluate our results by simulating 2, 3, ..., 99, and 100 queues for p=0.1, 0.2, ...0.8, and 0.9, each for four different HFJ cases, where rho = lambda / mu and mu is the average task service rate for a server. The maximum absolute offset for our hybrid solutions from all the simulations is less than 0.33 percent (1/300), which is a reasonable error ratio for simulation. The maximum offset for our upper bounds over all the simulations is 21 percent A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K/spl ges/2 . The dynamic-bubblesort model dynamically sorts the branches of the queues based on the number of the tasks waiting for synchronization in each branch. Jobs arrive with mean rate /spl lambda/ and a general arrival distribution. Upon arrival, a job forks into K tasks. Task k, k=1, 2, ..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite capacity queue. A job leaves the HFJ system as soon as all its tasks complete their service. In other words, tasks corresponding to the same job are joined before departing the HFJ system. We obtain a general and simple hybrid solution which combines analysis and simulation for the mean response time that we denote by T/sub K/. We obtain a very simple (as a function of T/sub 1/ and T/sub 2/ only) and general upper bound expression for T/sub K/ and we get an exact relationship between the cases for K=2 and 3. We evaluate our results by simulating 2, 3, ..., 99, and 100 queues for p=0.1, 0.2, ...0.8, and 0.9, each for four different HFJ cases, where /spl rho/=/spl lambda///spl mu/ and /spl mu/ is the average task service rate for a server. The maximum absolute offset for our hybrid solutions from all the simulations is less than 0.33 percent (1/300), which is a reasonable error ratio for simulation. The maximum offset for our upper bounds over all the simulations is 21 percent. |
Author | Ray Jinzhu Chen |
Author_xml | – sequence: 1 givenname: Ray surname: Chen middlename: Jinzhu fullname: Chen, Ray Jinzhu |
BookMark | eNqN0TtPwzAQAGALFYm2MLAyRQwghrS--BF7rCpeUiUWmCPXsdWU1C52M5Rfj0sqBgbE4rPuPt3pdCM0cN4ZhC4BTwCwnJYwkZRzJk_QEBgTeQGCDNIfU5bLAuQZGsW4xhgow3SIZrNstV-Gps6ib7td413mbWZ9eJ-ufeOyuHd6FbxrPtV3MaW2Kqi2NW320ZnOxHN0alUbzcUxjtHbw_3r_ClfvDw-z2eLXBNe7nJYgrASSy4oE8JgYgxlZV1SWutaKqmItFZzwYpSG83BKkZBsxqnB1PKyBjd9n23wafJcVdtmqhN2ypnfBcrCWlvXlCe5M2fshAEKIZ_QF4ywvhh9vUvuPZdcGndSkrCJCkFTuiuRzr4GIOx1TY0GxX2FeDqcJyqhKo_TrJXvW2MMT_uWPwCR0KJlw |
CODEN | ITDSEO |
CitedBy_id | crossref_primary_10_1109_TPDS_2010_168 crossref_primary_10_1016_j_jss_2015_08_036 crossref_primary_10_1016_j_peva_2010_03_002 crossref_primary_10_1017_S000186780001315X crossref_primary_10_1016_j_peva_2004_10_005 crossref_primary_10_3390_pr11020407 crossref_primary_10_1239_aap_1093962238 crossref_primary_10_1109_TPDS_2020_2982137 crossref_primary_10_1017_S0269964810000112 crossref_primary_10_1145_3300142 crossref_primary_10_1109_TPDS_2013_70 crossref_primary_10_1007_s10479_008_0312_7 crossref_primary_10_1016_j_peva_2017_04_005 |
Cites_doi | 10.1109/12.76416 10.2307/3214417 10.1109/TC.1982.1675926 10.1137/0145052 10.1109/32.4676 10.1109/12.477243 10.1109/12.16501 10.1109/12.40849 10.1145/65950.65957 10.1109/TC.1983.1676125 10.1137/0144074 10.1109/71.246075 10.1109/71.674321 10.1109/12.2213 10.2307/1427640 10.1109/TSE.1985.231547 10.1109/WSC.1994.717194 |
ContentType | Journal Article |
Copyright | Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2001 |
Copyright_xml | – notice: Copyright The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2001 |
DBID | RIA RIE AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D F28 FR3 |
DOI | 10.1109/71.946659 |
DatabaseName | IEEE All-Society Periodicals Package (ASPP) 1998-Present IEEE/IET Electronic Library (IEL) CrossRef Computer and Information Systems Abstracts Electronics & Communications Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional ANTE: Abstracts in New Technology & Engineering Engineering Research Database |
DatabaseTitle | CrossRef Technology Research Database Computer and Information Systems Abstracts – Academic Electronics & Communications Abstracts ProQuest Computer Science Collection Computer and Information Systems Abstracts Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Professional Engineering Research Database ANTE: Abstracts in New Technology & Engineering |
DatabaseTitleList | Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts Technology Research Database |
Database_xml | – sequence: 1 dbid: RIE name: IEEE url: https://proxy.k.utb.cz/login?url=https://ieeexplore.ieee.org/ sourceTypes: Publisher |
DeliveryMethod | fulltext_linktorsrc |
Discipline | Engineering Computer Science |
EISSN | 1558-2183 |
EndPage | 845 |
ExternalDocumentID | 2631938511 10_1109_71_946659 946659 |
GroupedDBID | --Z -~X .DC 0R~ 29I 4.4 5GY 5VS 6IK 97E AAJGR AASAJ AAYOK ABFSI ABQJQ ABVLG ACGFO ACIWK AENEX AETIX AI. AIBXA AKJIK ALLEH ALMA_UNASSIGNED_HOLDINGS ASUFR ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ CS3 DU5 E.L EBS EJD HZ~ H~9 ICLAB IEDLZ IFIPE IFJZH IPLJI JAVBF LAI M43 MS~ O9- OCL P2P PQQKQ RIA RIC RIE RIG RNI RNS RZB TN5 TWZ UHB VH1 XFK AAYXX CITATION 7SC 7SP 8FD JQ2 L7M L~C L~D F28 FR3 |
ID | FETCH-LOGICAL-c367t-1b18f909684588e03ee457d744dcd9a9a39ffc68527cec61fa541c5d01c504453 |
IEDL.DBID | RIE |
ISSN | 1045-9219 |
IngestDate | Fri Aug 16 10:52:41 EDT 2024 Fri Aug 16 06:03:52 EDT 2024 Fri Aug 16 07:50:23 EDT 2024 Thu Oct 10 19:07:43 EDT 2024 Fri Aug 23 03:08:07 EDT 2024 Wed Jun 26 19:28:25 EDT 2024 |
IsPeerReviewed | true |
IsScholarly | true |
Issue | 8 |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-c367t-1b18f909684588e03ee457d744dcd9a9a39ffc68527cec61fa541c5d01c504453 |
Notes | ObjectType-Article-2 SourceType-Scholarly Journals-1 ObjectType-Feature-1 content type line 23 |
PQID | 993593780 |
PQPubID | 23500 |
PageCount | 17 |
ParticipantIDs | proquest_miscellaneous_26753565 proquest_journals_993593780 crossref_primary_10_1109_71_946659 proquest_miscellaneous_914666246 proquest_miscellaneous_28314016 ieee_primary_946659 |
PublicationCentury | 2000 |
PublicationDate | 2001-08-01 |
PublicationDateYYYYMMDD | 2001-08-01 |
PublicationDate_xml | – month: 08 year: 2001 text: 2001-08-01 day: 01 |
PublicationDecade | 2000 |
PublicationPlace | New York |
PublicationPlace_xml | – name: New York |
PublicationTitle | IEEE transactions on parallel and distributed systems |
PublicationTitleAbbrev | TPDS |
PublicationYear | 2001 |
Publisher | IEEE The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
Publisher_xml | – name: IEEE – name: The Institute of Electrical and Electronics Engineers, Inc. (IEEE) |
References | kleinrock (bibL082912) 1975 lipsky (bibL082916) 1992 chen (bibL08294) 2000 bibL082910 bibL082922 bibL082923 sholl (bibL082924) 1992 bibL08299 bibL08298 bibL08297 chen (bibL08295) 1991 chen (bibL08293) 1998 bibL08292 bibL08291 bibL082921 bibL082911 lui (bibL082918) 1995; 44 bibL082913 bibL082914 bibL082915 bibL082917 bibL082919 ding (bibL08296) 1990 thomasian (bibL082925) 1994 makowski (bibL082920) 1994 |
References_xml | – ident: bibL082917 doi: 10.1109/12.76416 – ident: bibL082913 doi: 10.2307/3214417 – year: 2000 ident: bibL08294 article-title: 100-Queue Simulation Report I for `Hybrid Solution of Fork/Join Synchronization in Parallel Queues' contributor: fullname: chen – ident: bibL08299 doi: 10.1109/TC.1982.1675926 – ident: bibL08298 doi: 10.1137/0145052 – year: 1992 ident: bibL082916 publication-title: Queueing Theorem A Linear Algebraic Approach contributor: fullname: lipsky – ident: bibL082923 doi: 10.1109/32.4676 – volume: 44 start-page: 1371 year: 1995 ident: bibL082918 article-title: Bounding the Response Time of a Minimum Expected Delay Routing System: An Algorithmic Approach publication-title: IEEE Trans Computers doi: 10.1109/12.477243 contributor: fullname: lui – start-page: 145 year: 1994 ident: bibL082920 article-title: Interpolation Approximations for Symmetric Fork-Join Queues publication-title: Performance Evaluation 20 contributor: fullname: makowski – ident: bibL082911 doi: 10.1109/12.16501 – ident: bibL082921 doi: 10.1109/12.40849 – ident: bibL08292 doi: 10.1145/65950.65957 – ident: bibL082910 doi: 10.1109/TC.1983.1676125 – ident: bibL08297 doi: 10.1137/0144074 – ident: bibL082915 doi: 10.1109/71.246075 – year: 1998 ident: bibL08293 article-title: Analysis and Simulation of Parallel Fork-Join System contributor: fullname: chen – year: 1991 ident: bibL08295 article-title: Design and Modeling Policies for Two Server Fork/Join Queueing Systems contributor: fullname: chen – ident: bibL082919 doi: 10.1109/71.674321 – ident: bibL082922 doi: 10.1109/12.2213 – year: 1975 ident: bibL082912 publication-title: Queueing Systems vol I Theory contributor: fullname: kleinrock – ident: bibL08291 doi: 10.2307/1427640 – ident: bibL082914 doi: 10.1109/TSE.1985.231547 – start-page: 1 year: 1992 ident: bibL082924 article-title: Real-Time N-Queue Dynamic Schedule for M/G/1 System contributor: fullname: sholl – start-page: 361 year: 1994 ident: bibL082925 article-title: Approximate Solutions for M/G/1 Fork/Join Synchronization publication-title: Proc 1994 Winter Simulation Conf doi: 10.1109/WSC.1994.717194 contributor: fullname: thomasian – year: 1990 ident: bibL08296 article-title: On the Departure Process of a Fork-Join Queueing System publication-title: Proc ISMM Int'l Symp Parallel and Distributed Computing and Systems contributor: fullname: ding |
SSID | ssj0014504 |
Score | 1.810302 |
Snippet | A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K/spl... Task k, k=1, 2, ..., K, is assigned to the kth queuing system, which is a first-in-first-out server with a general service distribution and an infinite... A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K... A new analysis technique, dynamic-bubblesort analysis, is introduced for general K-queue first-in-first-out HFJ (homogenous fork/join queuing) systems of K 2 .... |
SourceID | proquest crossref ieee |
SourceType | Aggregation Database Publisher |
StartPage | 829 |
SubjectTerms | Analytical models Application software Arrivals Computational modeling Computer Society Concurrent computing Delay Mathematical models Parallel processing Performance analysis Queueing analysis Queues Queuing theory Servers Simulation Studies Synchronization Tasks Upper bound Upper bounds |
Title | A hybrid solution of fork/join synchronization in parallel queues |
URI | https://ieeexplore.ieee.org/document/946659 https://www.proquest.com/docview/993593780 https://search.proquest.com/docview/26753565 https://search.proquest.com/docview/28314016 https://search.proquest.com/docview/914666246 |
Volume | 12 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1LS8QwEA7qSQ8-VsX1GcRr16TN87iIIoKeFLyVNEnxsbTi7h7WX-8kbRffeCklTUuSziQzyXzfIHTirLEZGAoJzJBlwrgliS5Smihr09Iq4kxEyF3fiMs7dnXP71ue7YiF8d7H4DM_CLfxLN_Vdhq2yk4DFzrXi2hRkbSBas0PDBiPmQLBueCJBi1sSYQo0aeSDpoXPy09MZfKtwk4rioXaw1cexzJCEMwyfNgOikG9u0LVeM_G7yOVlvrEg8bcdhAC77qobUucwNuFbmHVj7QEG6i4RA_zAJyC3eCiOsSlyE6_Kl-rPB4VtnIodtANjEUBcbw0ciPMPQK-rWF7i7Ob88ukza1QmIzIScJLagqNbgvKiBVPcm8Z1w6yZizThttMl2WViieSuutoKXhjFruCFwIYzzbRktVXfkdhMEecGDGSCmcYU4xlRpZyIxSZ4iAKaSPjrtRz18aBo08eh5E55LmzRD1US-M3rxCV7rX_Z68Va1xrgOWOJMKPnw0fwo6EQ46TOXr6ThPwQsKAvhHDZUFz1L0Ef6lhoYlRIiUid0fm7aHlptotBAKuI-WJq9TfwDmyaQ4jIL5DvnU4wY |
link.rule.ids | 315,786,790,802,27955,27956,55107 |
linkProvider | IEEE |
linkToHtml | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwjV1LT9wwEB5ROLQcoN22YoEWq-o1i534eVyhom15nEDiFjm2oz5WScXuHuDXM3aSFX1Q9RJFiRPZzoxnJp7vG4CP3llXoKOQ4QpZZ1w4mpkqZ5l2Lq-dpt4mhNzFpZxd8y834qbn2U5YmBBCSj4Lk3ia9vJ961bxV9lx5EIX5hlsoZmnqgNrrbcMuEi1AjG8EJlBPexphBg1x4pNukd_MT6pmsofS3CyK6e7HWB7kegIYzrJj8lqWU3c_W9kjf_Z5Zew0_uXZNoJxCvYCM0IdofaDaRX5RFsPyIifA3TKfl6F7FbZBBF0takjvnh39tvDVncNS6x6HagTYKXImf4fB7mBEeF43oD16efrk5mWV9cIXOFVMuMVUzXBgMYHbGqgRYhcKG84tw7b6yxhalrJ7XIlQtOstoKzpzwFA-Uc1G8hc2mbcIeEPQIPDoySklvuddc51ZVqmDMWypxERnDh2HWy58dh0aZYg9qSsXKborGMIqzt24wXD0YPk_ZK9eiNBFNXCiNLz5a30WtiFsdtgntalHmGAdFEfxHC13E2FKOgTzRwqARkTLncv-vXTuC57Ori_Py_PPl2QG86HLTYmLgIWwub1fhHTory-p9EtIHFH7mWg |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+hybrid+solution+of+fork%2Fjoin+synchronization+in+parallel+queues&rft.jtitle=IEEE+transactions+on+parallel+and+distributed+systems&rft.au=Chen%2C+Ray+Jinzhu&rft.date=2001-08-01&rft.issn=1045-9219&rft.volume=12&rft.issue=8&rft.spage=829&rft.epage=845&rft_id=info:doi/10.1109%2F71.946659&rft.externalDBID=NO_FULL_TEXT |
thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1045-9219&client=summon |
thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1045-9219&client=summon |
thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1045-9219&client=summon |