Academic Journal

Distributed Sketching Lower Bounds for k-Edge Connected Spanning Subgraphs, BFS Trees, and LCL Problems

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Distributed Sketching Lower Bounds for k-Edge Connected Spanning Subgraphs, BFS Trees, and LCL Problems
Συγγραφείς: Robinson, Peter
Συνεισφορές: Peter Robinson
Στοιχεία εκδότη: Schloss Dagstuhl – Leibniz-Zentrum für Informatik
Έτος έκδοσης: 2023
Συλλογή: DROPS - Dagstuhl Research Online Publication Server (Schloss Dagstuhl - Leibniz Center for Informatics )
Θεματικοί όροι: Distributed graph algorithm, graph sketching, streaming
Περιγραφή: We investigate graph problems in the distributed sketching model, where each node sends a single message to a referee who computes the output. We define a class of graphs and give a framework for proving lower bounds for certain embeddable problems, which leads to several new results: First, we present a tight lower bound of Ω(n) bits for the message size of computing a breadth-first search (BFS) tree. For locally-checkable labeling (LCL) problems, we show that verifying whether a given vertex labeling forms a weak 2-coloring requires messages of Ω(n^{1/3}log^{2/3}n) bits, and the same lower bound holds for verifying whether a subset of nodes forms a maximal independent set. We also prove that computing a k-edge connected spanning subgraph (k-ECSS) requires messages of size at least Ω(klog²(n/k)), which is tight up to a logarithmic factor. To show these results, we define a simultaneous multiparty (SMP) model of communication complexity, where the players obtain certain subgraphs as their input, and develop a generic embedding argument that allows us to prove lower bounds for the graph sketching model by using reductions from the SMP model. We point out that these results also extend to single-round algorithms in the broadcast congested clique. We also (nearly) settle the space complexity of the k-ECSS problem in the streaming model by extending the work of Kapralov, Nelson, Pachoki, Wang, and Woodruff (FOCS 2017): We prove a communication complexity lower bound for a direct sum variant of the UR^⊂_k problem and show that this implies Ω(k nlog²(n/k)) bits of memory for computing a k-ECSS. This is known to be optimal up to a logarithmic factor.
Τύπος εγγράφου: article in journal/newspaper
conference object
Περιγραφή αρχείου: application/pdf
Γλώσσα: English
Relation: Is Part Of LIPIcs, Volume 281, 37th International Symposium on Distributed Computing (DISC 2023); https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.DISC.2023.32
DOI: 10.4230/LIPIcs.DISC.2023.32
Διαθεσιμότητα: https://doi.org/10.4230/LIPIcs.DISC.2023.32
https://nbn-resolving.org/urn:nbn:de:0030-drops-191589
https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.DISC.2023.32
Rights: https://creativecommons.org/licenses/by/4.0/legalcode
Αριθμός Καταχώρησης: edsbas.51649566
Βάση Δεδομένων: BASE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://doi.org/10.4230/LIPIcs.DISC.2023.32#
    Name: EDS - BASE (ns324271)
    Category: fullText
    Text: View record from BASE
Header DbId: edsbas
DbLabel: BASE
An: edsbas.51649566
RelevancyScore: 944
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 944.480346679688
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Distributed Sketching Lower Bounds for k-Edge Connected Spanning Subgraphs, BFS Trees, and LCL Problems
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Robinson%2C+Peter%22">Robinson, Peter</searchLink>
– Name: Author
  Label: Contributors
  Group: Au
  Data: Peter Robinson
– Name: Publisher
  Label: Publisher Information
  Group: PubInfo
  Data: Schloss Dagstuhl – Leibniz-Zentrum für Informatik
– Name: DatePubCY
  Label: Publication Year
  Group: Date
  Data: 2023
– Name: Subset
  Label: Collection
  Group: HoldingsInfo
  Data: DROPS - Dagstuhl Research Online Publication Server (Schloss Dagstuhl - Leibniz Center for Informatics )
– Name: Subject
  Label: Subject Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Distributed+graph+algorithm%22">Distributed graph algorithm</searchLink><br /><searchLink fieldCode="DE" term="%22graph+sketching%22">graph sketching</searchLink><br /><searchLink fieldCode="DE" term="%22streaming%22">streaming</searchLink>
– Name: Abstract
  Label: Description
  Group: Ab
  Data: We investigate graph problems in the distributed sketching model, where each node sends a single message to a referee who computes the output. We define a class of graphs and give a framework for proving lower bounds for certain embeddable problems, which leads to several new results: First, we present a tight lower bound of Ω(n) bits for the message size of computing a breadth-first search (BFS) tree. For locally-checkable labeling (LCL) problems, we show that verifying whether a given vertex labeling forms a weak 2-coloring requires messages of Ω(n^{1/3}log^{2/3}n) bits, and the same lower bound holds for verifying whether a subset of nodes forms a maximal independent set. We also prove that computing a k-edge connected spanning subgraph (k-ECSS) requires messages of size at least Ω(klog²(n/k)), which is tight up to a logarithmic factor. To show these results, we define a simultaneous multiparty (SMP) model of communication complexity, where the players obtain certain subgraphs as their input, and develop a generic embedding argument that allows us to prove lower bounds for the graph sketching model by using reductions from the SMP model. We point out that these results also extend to single-round algorithms in the broadcast congested clique. We also (nearly) settle the space complexity of the k-ECSS problem in the streaming model by extending the work of Kapralov, Nelson, Pachoki, Wang, and Woodruff (FOCS 2017): We prove a communication complexity lower bound for a direct sum variant of the UR^⊂_k problem and show that this implies Ω(k nlog²(n/k)) bits of memory for computing a k-ECSS. This is known to be optimal up to a logarithmic factor.
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: article in journal/newspaper<br />conference object
– Name: Format
  Label: File Description
  Group: SrcInfo
  Data: application/pdf
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: NoteTitleSource
  Label: Relation
  Group: SrcInfo
  Data: Is Part Of LIPIcs, Volume 281, 37th International Symposium on Distributed Computing (DISC 2023); https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.DISC.2023.32
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.4230/LIPIcs.DISC.2023.32
– Name: URL
  Label: Availability
  Group: URL
  Data: https://doi.org/10.4230/LIPIcs.DISC.2023.32<br />https://nbn-resolving.org/urn:nbn:de:0030-drops-191589<br />https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.DISC.2023.32
– Name: Copyright
  Label: Rights
  Group: Cpyrght
  Data: https://creativecommons.org/licenses/by/4.0/legalcode
– Name: AN
  Label: Accession Number
  Group: ID
  Data: edsbas.51649566
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsbas&AN=edsbas.51649566
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.4230/LIPIcs.DISC.2023.32
    Languages:
      – Text: English
    Subjects:
      – SubjectFull: Distributed graph algorithm
        Type: general
      – SubjectFull: graph sketching
        Type: general
      – SubjectFull: streaming
        Type: general
    Titles:
      – TitleFull: Distributed Sketching Lower Bounds for k-Edge Connected Spanning Subgraphs, BFS Trees, and LCL Problems
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Robinson, Peter
      – PersonEntity:
          Name:
            NameFull: Peter Robinson
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 01
              Type: published
              Y: 2023
          Identifiers:
            – Type: issn-locals
              Value: edsbas
            – Type: issn-locals
              Value: edsbas.oa
ResultId 1