Endre søk
Link to record
Permanent link

Direct link
Publikasjoner (4 av 4) Visa alla publikasjoner
Bilardi, G., Scquizzato, M. & Silvestri, F. (2018). A Lower Bound Technique for Communication in BSP. ACM TRANSACTIONS ON PARALLEL COMPUTING, 4(3), Article ID UNSP 14.
Åpne denne publikasjonen i ny fane eller vindu >>A Lower Bound Technique for Communication in BSP
2018 (engelsk)Inngår i: ACM TRANSACTIONS ON PARALLEL COMPUTING, ISSN 2329-4949, Vol. 4, nr 3, artikkel-id UNSP 14Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

Communication is a major factor determining the performance of algorithms on current computing systems; it is therefore valuable to provide tight lower bounds on the communication complexity of computations. This article presents a lower bound technique for the communication complexity in the bulk-synchronous parallel (BSP) model of a given class of DAG computations. The derived bound is expressed in terms of the switching potential of a DAG, that is, the number of permutations that the DAG can realize when viewed as a switching network. The proposed technique yields tight lower bounds for the fast Fourier transform (FFT), and for any sorting and permutation network. A stronger bound is also derived for the periodic balanced sorting network, by applying this technique to suitable subnetworks. Finally, we demonstrate that the switching potential captures communication requirements even in computational models different from BSP, such as the I/O model and the LPRAM.

sted, utgiver, år, opplag, sider
ASSOC COMPUTING MACHINERY, 2018
Emneord
Communication, parallel computing, lower bounds, switching networks, FFT, sorting networks
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-231235 (URN)10.1145/3181776 (DOI)000434643100003 ()2-s2.0-85054891846 (Scopus ID)
Merknad

QC 20180627

Tilgjengelig fra: 2018-06-27 Laget: 2018-06-27 Sist oppdatert: 2022-06-26bibliografisk kontrollert
Pandurangan, G., Robinson, P. & Scquizzato, M. (2018). Fast distributed algorithms for connectivity and MST in Large Graphs. ACM Transactions on Parallel Computing, 5(1), 1-22, Article ID a4.
Åpne denne publikasjonen i ny fane eller vindu >>Fast distributed algorithms for connectivity and MST in Large Graphs
2018 (engelsk)Inngår i: ACM Transactions on Parallel Computing, ISSN 2329-4949, Vol. 5, nr 1, s. 1-22, artikkel-id a4Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

Motivated by the increasing need to understand the algorithmic foundations of distributed large-scale graph computations, we study a number of fundamental graph problems in amessage-passing model for distributed computing where k ≥ 2 machines jointly perform computations on graphs with n nodes (typically, n > k). The input graph is assumed to be initially randomly partitioned among the k machines, a common implementation in many real-world systems. Communication is point-To-point, and the goal is to minimize the number of communication rounds of the computation. Our main result is an (almost) optimal distributed randomized algorithm for graph connectivity. Our algorithm runs in Õ (n/k2) rounds ( Õ notation hides a polylog(n) factor and an additive polylog(n) term). This improves over the best previously known bound of Õ (n/k) [Klauck et al., SODA 2015] and is optimal (up to a polylogarithmic factor) in light of an existing lower bound of ω(n/k2). Our improved algorithm uses a bunch of techniques, including linear graph sketching, that prove useful in the design of efficient distributed graph algorithms. Using the connectivity algorithm as a building block, we then present fast randomized algorithms for computing minimum spanning trees, (approximate) min-cuts, and for many graph verification problems. All these algorithms take Õ (n/k2) rounds and are optimal up to polylogarithmic factors. We also show an almost matching lower bound of ω (n/k2) rounds for many graph verification problems by leveraging lower bounds in random-partition communication complexity. 

sted, utgiver, år, opplag, sider
Association for Computing Machinery (ACM), 2018
Emneord
Distributed graph algorithms, Graph connectivity, Graph sketching, Minimum spanning trees, Distributed computer systems, Algorithmic foundations, Communication complexity, Connectivity algorithms, Poly-logarithmic factors, Trees (mathematics)
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-314288 (URN)10.1145/3209689 (DOI)000455921800004 ()2-s2.0-85051102207 (Scopus ID)
Merknad

QC 20220620

Tilgjengelig fra: 2022-06-20 Laget: 2022-06-20 Sist oppdatert: 2022-06-25bibliografisk kontrollert
Pandurangan, G., Robinson, P. & Scquizzato, M. (2018). On the distributed complexity of large-scale graph computations. In: SPAA '18 Proceedings of the 30th on Symposium on Parallelism in Algorithms and Architectures: . Paper presented at 30th ACM Symposium on Parallelism in Algorithms and Architectures, SPAA 2018, Vienna, Austria, 16 July 2018 through 18 July 2018 (pp. 405-414). Association for Computing Machinery (ACM)
Åpne denne publikasjonen i ny fane eller vindu >>On the distributed complexity of large-scale graph computations
2018 (engelsk)Inngår i: SPAA '18 Proceedings of the 30th on Symposium on Parallelism in Algorithms and Architectures, Association for Computing Machinery (ACM), 2018, s. 405-414Konferansepaper, Publicerat paper (Fagfellevurdert)
Abstract [en]

Motivated by the increasing need to understand the distributed algorithmic foundations of large-scale graph computations, we study some fundamental graph problems in a message-passing model for distributed computing where k 2 machines jointly perform computations on graphs with n nodes (typically, n ≫ k). The input graph is assumed to be initially randomly partitioned among the k machines, a common implementation in many real-world systems. Communication is point-to-point, and the goal is to minimize the number of communication rounds of the computation.

sted, utgiver, år, opplag, sider
Association for Computing Machinery (ACM), 2018
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-234110 (URN)10.1145/3210377.3210409 (DOI)000545269600048 ()2-s2.0-85051131123 (Scopus ID)9781450357999 (ISBN)
Konferanse
30th ACM Symposium on Parallelism in Algorithms and Architectures, SPAA 2018, Vienna, Austria, 16 July 2018 through 18 July 2018
Merknad

QC 20180905

Tilgjengelig fra: 2018-09-05 Laget: 2018-09-05 Sist oppdatert: 2022-06-26bibliografisk kontrollert
Schmid, S., Pandurangan, G., Robinson, P. & Scquizzato, M. (2018). The Distributed Minimum Spanning Tree Problem. Bulletin of the European Association for Theoretical Computer Science (125), 51-80
Åpne denne publikasjonen i ny fane eller vindu >>The Distributed Minimum Spanning Tree Problem
2018 (engelsk)Inngår i: Bulletin of the European Association for Theoretical Computer Science, ISSN 0252-9742, nr 125, s. 51-80Artikkel i tidsskrift (Fagfellevurdert) Published
Abstract [en]

This article surveys the distributed minimum spanning tree (MST) problem, a central and one of the most studied problems in distributed computing. In this problem, we are given a network, represented as a weighted graph G = (v, E), and the nodes in the network communicate by message passing via the edges of G with the goal of constructing an MST of G in a distributed fashion, i.e., each node should identify the MST edges incident to itself. This article summarizes the long line of research in designing efficient distributed algorithms and showing lower bounds for the distributed MST problem, including the most recent developments which have focused on algorithms that are simultaneously round- and message-optimal.

sted, utgiver, år, opplag, sider
EUROPEAN ASSOC THEORETICAL COMPUTER SCIENCE, 2018
HSV kategori
Identifikatorer
urn:nbn:se:kth:diva-232263 (URN)000436902200002 ()
Merknad

QC 20180719

Tilgjengelig fra: 2018-07-19 Laget: 2018-07-19 Sist oppdatert: 2022-06-26bibliografisk kontrollert
Organisasjoner
Identifikatorer
ORCID-id: ORCID iD iconorcid.org/0000-0002-9108-2448