The full story of 1000 cores View Full Text


Ontology type: schema:ScholarlyArticle      Open Access: True


Article Info

DATE

2022-04-29

AUTHORS

Tiemo Bang, Norman May, Ilia Petrov, Carsten Binnig

ABSTRACT

In our initial DaMoN paper, we set out the goal to revisit the results of “Starring into the Abyss [...] of Concurrency Control with [1000] Cores” (Yu in Proc. VLDB Endow 8: 209-220, 2014). Against their assumption, today we do not see single-socket CPUs with 1000 cores. Instead, multi-socket hardware is prevalent today and in fact offers over 1000 cores. Hence, we evaluated concurrency control (CC) schemes on a real (Intel-based) multi-socket platform. To our surprise, we made interesting findings opposing results of the original analysis that we discussed in our initial DaMoN paper. In this paper, we further broaden our analysis, detailing the effect of hardware and workload characteristics via additional real hardware platforms (IBM Power8 and 9) and the full TPC-C transaction mix. Among others, we identified clear connections between the performance of the CC schemes and hardware characteristics, especially concerning NUMA and CPU cache. Overall, we conclude that no CC scheme can efficiently make use of large multi-socket hardware in a robust manner and suggest several directions on how CC schemes and overall OLTP DBMS should evolve in future. More... »

PAGES

1-29

References to SciGraph publications

  • 2019-08-13. Avoiding Scalability Collapse by Restricting Concurrency in EURO-PAR 2019: PARALLEL PROCESSING
  • 2019-04-24. Adaptive Transaction Scheduling for Highly Contended Workloads in DATABASE SYSTEMS FOR ADVANCED APPLICATIONS
  • 2015-12-29. Characterization of the Impact of Hardware Islands on OLTP in THE VLDB JOURNAL
  • Identifiers

    URI

    http://scigraph.springernature.com/pub.10.1007/s00778-022-00742-4

    DOI

    http://dx.doi.org/10.1007/s00778-022-00742-4

    DIMENSIONS

    https://app.dimensions.ai/details/publication/pub.1147508792


    Indexing Status Check whether this publication has been indexed by Scopus and Web Of Science using the SN Indexing Status Tool
    Incoming Citations Browse incoming citations for this publication using opencitations.net

    JSON-LD is the canonical representation for SciGraph data.

    TIP: You can open this SciGraph record using an external JSON-LD service: JSON-LD Playground Google SDTT

    [
      {
        "@context": "https://springernature.github.io/scigraph/jsonld/sgcontext.json", 
        "about": [
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/08", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Information and Computing Sciences", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0804", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Data Format", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0805", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Distributed Computing", 
            "type": "DefinedTerm"
          }, 
          {
            "id": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/0806", 
            "inDefinedTermSet": "http://purl.org/au-research/vocabulary/anzsrc-for/2008/", 
            "name": "Information Systems", 
            "type": "DefinedTerm"
          }
        ], 
        "author": [
          {
            "affiliation": {
              "alternateName": "Technical University of Darmstadt & SAP SE, Darmstadt, Germany", 
              "id": "http://www.grid.ac/institutes/grid.6546.1", 
              "name": [
                "Technical University of Darmstadt & SAP SE, Darmstadt, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Bang", 
            "givenName": "Tiemo", 
            "id": "sg:person.012034054161.09", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012034054161.09"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "SAP SE, Walldorf, Germany", 
              "id": "http://www.grid.ac/institutes/grid.19008.30", 
              "name": [
                "SAP SE, Walldorf, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "May", 
            "givenName": "Norman", 
            "id": "sg:person.012077122055.34", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012077122055.34"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Reutlingen University, Reutlingen, Germany", 
              "id": "http://www.grid.ac/institutes/grid.434088.3", 
              "name": [
                "Reutlingen University, Reutlingen, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Petrov", 
            "givenName": "Ilia", 
            "id": "sg:person.07400607751.62", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07400607751.62"
            ], 
            "type": "Person"
          }, 
          {
            "affiliation": {
              "alternateName": "Technical University of Darmstadt, Darmstadt, Germany", 
              "id": "http://www.grid.ac/institutes/grid.6546.1", 
              "name": [
                "Technical University of Darmstadt, Darmstadt, Germany"
              ], 
              "type": "Organization"
            }, 
            "familyName": "Binnig", 
            "givenName": "Carsten", 
            "id": "sg:person.011106101322.26", 
            "sameAs": [
              "https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011106101322.26"
            ], 
            "type": "Person"
          }
        ], 
        "citation": [
          {
            "id": "sg:pub.10.1007/978-3-030-29400-7_26", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1120394625", 
              "https://doi.org/10.1007/978-3-030-29400-7_26"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/s00778-015-0413-2", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1029608657", 
              "https://doi.org/10.1007/s00778-015-0413-2"
            ], 
            "type": "CreativeWork"
          }, 
          {
            "id": "sg:pub.10.1007/978-3-030-18590-9_90", 
            "sameAs": [
              "https://app.dimensions.ai/details/publication/pub.1113642328", 
              "https://doi.org/10.1007/978-3-030-18590-9_90"
            ], 
            "type": "CreativeWork"
          }
        ], 
        "datePublished": "2022-04-29", 
        "datePublishedReg": "2022-04-29", 
        "description": "In our initial DaMoN paper, we set out the goal to revisit the results of \u201cStarring into the Abyss [...] of Concurrency Control with [1000] Cores\u201d (Yu in Proc. VLDB Endow 8: 209-220, 2014). Against their assumption, today we do not see single-socket CPUs with 1000 cores. Instead, multi-socket hardware is prevalent today and in fact offers over 1000 cores. Hence, we evaluated concurrency control (CC) schemes on a real (Intel-based) multi-socket platform. To our surprise, we made interesting findings opposing results of the original analysis that we discussed in our initial DaMoN paper. In this paper, we further broaden our analysis, detailing the effect of hardware and workload characteristics via additional real hardware platforms (IBM Power8 and 9) and the full TPC-C transaction mix. Among others, we identified clear connections between the performance of the CC schemes and hardware characteristics, especially concerning NUMA and CPU cache. Overall, we conclude that no CC scheme can efficiently make use of large multi-socket hardware in a robust manner and suggest several directions on how CC schemes and overall OLTP DBMS should evolve in future.", 
        "genre": "article", 
        "id": "sg:pub.10.1007/s00778-022-00742-4", 
        "inLanguage": "en", 
        "isAccessibleForFree": true, 
        "isPartOf": [
          {
            "id": "sg:journal.1044889", 
            "issn": [
              "1066-8888", 
              "0949-877X"
            ], 
            "name": "The VLDB Journal", 
            "publisher": "Springer Nature", 
            "type": "Periodical"
          }
        ], 
        "keywords": [
          "concurrency control", 
          "multi-socket platforms", 
          "real hardware platform", 
          "CC scheme", 
          "effect of hardware", 
          "workload characteristics", 
          "hardware platform", 
          "transaction mix", 
          "hardware characteristics", 
          "CPU cache", 
          "hardware", 
          "robust manner", 
          "platform", 
          "scheme", 
          "CPU", 
          "prevalent today", 
          "cache", 
          "today", 
          "NuMA", 
          "goal", 
          "core", 
          "interesting finding", 
          "performance", 
          "dBm", 
          "results", 
          "control", 
          "connection", 
          "manner", 
          "assumption", 
          "fact", 
          "surprise", 
          "original analysis", 
          "analysis", 
          "characteristics", 
          "use", 
          "direction", 
          "future", 
          "findings", 
          "effect", 
          "mix", 
          "clear connection", 
          "full story", 
          "abyss", 
          "story", 
          "paper"
        ], 
        "name": "The full story of 1000 cores", 
        "pagination": "1-29", 
        "productId": [
          {
            "name": "dimensions_id", 
            "type": "PropertyValue", 
            "value": [
              "pub.1147508792"
            ]
          }, 
          {
            "name": "doi", 
            "type": "PropertyValue", 
            "value": [
              "10.1007/s00778-022-00742-4"
            ]
          }
        ], 
        "sameAs": [
          "https://doi.org/10.1007/s00778-022-00742-4", 
          "https://app.dimensions.ai/details/publication/pub.1147508792"
        ], 
        "sdDataset": "articles", 
        "sdDatePublished": "2022-06-01T22:23", 
        "sdLicense": "https://scigraph.springernature.com/explorer/license/", 
        "sdPublisher": {
          "name": "Springer Nature - SN SciGraph project", 
          "type": "Organization"
        }, 
        "sdSource": "s3://com-springernature-scigraph/baseset/20220601/entities/gbq_results/article/article_938.jsonl", 
        "type": "ScholarlyArticle", 
        "url": "https://doi.org/10.1007/s00778-022-00742-4"
      }
    ]
     

    Download the RDF metadata as:  json-ld nt turtle xml License info

    HOW TO GET THIS DATA PROGRAMMATICALLY:

    JSON-LD is a popular format for linked data which is fully compatible with JSON.

    curl -H 'Accept: application/ld+json' 'https://scigraph.springernature.com/pub.10.1007/s00778-022-00742-4'

    N-Triples is a line-based linked data format ideal for batch operations.

    curl -H 'Accept: application/n-triples' 'https://scigraph.springernature.com/pub.10.1007/s00778-022-00742-4'

    Turtle is a human-readable linked data format.

    curl -H 'Accept: text/turtle' 'https://scigraph.springernature.com/pub.10.1007/s00778-022-00742-4'

    RDF/XML is a standard XML format for linked data.

    curl -H 'Accept: application/rdf+xml' 'https://scigraph.springernature.com/pub.10.1007/s00778-022-00742-4'


     

    This table displays all metadata directly associated to this object as RDF triples.

    146 TRIPLES      22 PREDICATES      73 URIs      60 LITERALS      4 BLANK NODES

    Subject Predicate Object
    1 sg:pub.10.1007/s00778-022-00742-4 schema:about anzsrc-for:08
    2 anzsrc-for:0804
    3 anzsrc-for:0805
    4 anzsrc-for:0806
    5 schema:author N146e7889a08644379a14961821560055
    6 schema:citation sg:pub.10.1007/978-3-030-18590-9_90
    7 sg:pub.10.1007/978-3-030-29400-7_26
    8 sg:pub.10.1007/s00778-015-0413-2
    9 schema:datePublished 2022-04-29
    10 schema:datePublishedReg 2022-04-29
    11 schema:description In our initial DaMoN paper, we set out the goal to revisit the results of “Starring into the Abyss [...] of Concurrency Control with [1000] Cores” (Yu in Proc. VLDB Endow 8: 209-220, 2014). Against their assumption, today we do not see single-socket CPUs with 1000 cores. Instead, multi-socket hardware is prevalent today and in fact offers over 1000 cores. Hence, we evaluated concurrency control (CC) schemes on a real (Intel-based) multi-socket platform. To our surprise, we made interesting findings opposing results of the original analysis that we discussed in our initial DaMoN paper. In this paper, we further broaden our analysis, detailing the effect of hardware and workload characteristics via additional real hardware platforms (IBM Power8 and 9) and the full TPC-C transaction mix. Among others, we identified clear connections between the performance of the CC schemes and hardware characteristics, especially concerning NUMA and CPU cache. Overall, we conclude that no CC scheme can efficiently make use of large multi-socket hardware in a robust manner and suggest several directions on how CC schemes and overall OLTP DBMS should evolve in future.
    12 schema:genre article
    13 schema:inLanguage en
    14 schema:isAccessibleForFree true
    15 schema:isPartOf sg:journal.1044889
    16 schema:keywords CC scheme
    17 CPU
    18 CPU cache
    19 NuMA
    20 abyss
    21 analysis
    22 assumption
    23 cache
    24 characteristics
    25 clear connection
    26 concurrency control
    27 connection
    28 control
    29 core
    30 dBm
    31 direction
    32 effect
    33 effect of hardware
    34 fact
    35 findings
    36 full story
    37 future
    38 goal
    39 hardware
    40 hardware characteristics
    41 hardware platform
    42 interesting finding
    43 manner
    44 mix
    45 multi-socket platforms
    46 original analysis
    47 paper
    48 performance
    49 platform
    50 prevalent today
    51 real hardware platform
    52 results
    53 robust manner
    54 scheme
    55 story
    56 surprise
    57 today
    58 transaction mix
    59 use
    60 workload characteristics
    61 schema:name The full story of 1000 cores
    62 schema:pagination 1-29
    63 schema:productId Na68a14f4a96d41f0b23ea9fb3d067bf8
    64 Nf13d1f7fd05a4c3cbbcf48d657a005ef
    65 schema:sameAs https://app.dimensions.ai/details/publication/pub.1147508792
    66 https://doi.org/10.1007/s00778-022-00742-4
    67 schema:sdDatePublished 2022-06-01T22:23
    68 schema:sdLicense https://scigraph.springernature.com/explorer/license/
    69 schema:sdPublisher Nb6b84e45abcd4f25b692909d0b27a303
    70 schema:url https://doi.org/10.1007/s00778-022-00742-4
    71 sgo:license sg:explorer/license/
    72 sgo:sdDataset articles
    73 rdf:type schema:ScholarlyArticle
    74 N146e7889a08644379a14961821560055 rdf:first sg:person.012034054161.09
    75 rdf:rest Nc278c65c6c6e4fb38b3336838a673e57
    76 N4a4f84c0eece461f8ba39cd76668b970 rdf:first sg:person.011106101322.26
    77 rdf:rest rdf:nil
    78 N7a985996c5274b899c35788b4fd03963 rdf:first sg:person.07400607751.62
    79 rdf:rest N4a4f84c0eece461f8ba39cd76668b970
    80 Na68a14f4a96d41f0b23ea9fb3d067bf8 schema:name doi
    81 schema:value 10.1007/s00778-022-00742-4
    82 rdf:type schema:PropertyValue
    83 Nb6b84e45abcd4f25b692909d0b27a303 schema:name Springer Nature - SN SciGraph project
    84 rdf:type schema:Organization
    85 Nc278c65c6c6e4fb38b3336838a673e57 rdf:first sg:person.012077122055.34
    86 rdf:rest N7a985996c5274b899c35788b4fd03963
    87 Nf13d1f7fd05a4c3cbbcf48d657a005ef schema:name dimensions_id
    88 schema:value pub.1147508792
    89 rdf:type schema:PropertyValue
    90 anzsrc-for:08 schema:inDefinedTermSet anzsrc-for:
    91 schema:name Information and Computing Sciences
    92 rdf:type schema:DefinedTerm
    93 anzsrc-for:0804 schema:inDefinedTermSet anzsrc-for:
    94 schema:name Data Format
    95 rdf:type schema:DefinedTerm
    96 anzsrc-for:0805 schema:inDefinedTermSet anzsrc-for:
    97 schema:name Distributed Computing
    98 rdf:type schema:DefinedTerm
    99 anzsrc-for:0806 schema:inDefinedTermSet anzsrc-for:
    100 schema:name Information Systems
    101 rdf:type schema:DefinedTerm
    102 sg:journal.1044889 schema:issn 0949-877X
    103 1066-8888
    104 schema:name The VLDB Journal
    105 schema:publisher Springer Nature
    106 rdf:type schema:Periodical
    107 sg:person.011106101322.26 schema:affiliation grid-institutes:grid.6546.1
    108 schema:familyName Binnig
    109 schema:givenName Carsten
    110 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.011106101322.26
    111 rdf:type schema:Person
    112 sg:person.012034054161.09 schema:affiliation grid-institutes:grid.6546.1
    113 schema:familyName Bang
    114 schema:givenName Tiemo
    115 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012034054161.09
    116 rdf:type schema:Person
    117 sg:person.012077122055.34 schema:affiliation grid-institutes:grid.19008.30
    118 schema:familyName May
    119 schema:givenName Norman
    120 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.012077122055.34
    121 rdf:type schema:Person
    122 sg:person.07400607751.62 schema:affiliation grid-institutes:grid.434088.3
    123 schema:familyName Petrov
    124 schema:givenName Ilia
    125 schema:sameAs https://app.dimensions.ai/discover/publication?and_facet_researcher=ur.07400607751.62
    126 rdf:type schema:Person
    127 sg:pub.10.1007/978-3-030-18590-9_90 schema:sameAs https://app.dimensions.ai/details/publication/pub.1113642328
    128 https://doi.org/10.1007/978-3-030-18590-9_90
    129 rdf:type schema:CreativeWork
    130 sg:pub.10.1007/978-3-030-29400-7_26 schema:sameAs https://app.dimensions.ai/details/publication/pub.1120394625
    131 https://doi.org/10.1007/978-3-030-29400-7_26
    132 rdf:type schema:CreativeWork
    133 sg:pub.10.1007/s00778-015-0413-2 schema:sameAs https://app.dimensions.ai/details/publication/pub.1029608657
    134 https://doi.org/10.1007/s00778-015-0413-2
    135 rdf:type schema:CreativeWork
    136 grid-institutes:grid.19008.30 schema:alternateName SAP SE, Walldorf, Germany
    137 schema:name SAP SE, Walldorf, Germany
    138 rdf:type schema:Organization
    139 grid-institutes:grid.434088.3 schema:alternateName Reutlingen University, Reutlingen, Germany
    140 schema:name Reutlingen University, Reutlingen, Germany
    141 rdf:type schema:Organization
    142 grid-institutes:grid.6546.1 schema:alternateName Technical University of Darmstadt & SAP SE, Darmstadt, Germany
    143 Technical University of Darmstadt, Darmstadt, Germany
    144 schema:name Technical University of Darmstadt & SAP SE, Darmstadt, Germany
    145 Technical University of Darmstadt, Darmstadt, Germany
    146 rdf:type schema:Organization
     




    Preview window. Press ESC to close (or click here)


    ...