Definition
Bandwidth-latency models are a group of performance models for parallel programs that focus on modeling the communication between the processes in terms of network bandwidth and latency, allowing quite precise performance estimations. While originally developed for distributed-memory architectures, these models also apply to machines with nonuniform memory access (NUMA), like the modern multi-core architectures.
Discussion
Introduction
The foremost goal of parallel programming is to speed up algorithms that would be too slow when executed sequentially. Achieving this so-called speedup requires a deep understanding of the performance of the inter-process communication and synchronization, together with the algorithm’s computation. Both computation and communication/synchronization performance strongly depend on properties of the machine architecture in use. The strength of the bandwidth-latency models...
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Bibliography
Alexandrov A, Ionescu M, Schauser KE, Scheiman C (1995) LogGP: incorporating long messages into the LogP model – one step closer towards a realistic model for parallel computation. In: 7th ACM Symposium on Parallel Algorithms and Architectures (SPAA’95), Santa Barbara, California, pp 95–105, July 1995
Bilardi G, Herley KT, Pietracaprina A, Pucci G, Spirakis P (1999) BSP versus LogP. Algorithmica 24:405–422
Culler DE, Dusseau AC, Martin RP, Schauser KE (1994) Fast parallel sorting under LogP: from theory to practice. In: Portability and Performance for Parallel Processing, Wiley, Southampton, pp 71–98, 1994
Culler DE, Karp R, Sahay A, Schauser KE, Santos E, Subramonian R, von Eicken T (1993) LogP: towards a realistic model of parallel computation. In: 4th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP’93), pp 1–12, 1993
Goudreau MW, Hill JM, Lang K, McColl WF, Rao SD, Stefanescu DC, Suel T, Tsantilas T (1996) A proposal for a BSP Worldwide standard. Technical Report, BSP Worldwide, www.bsp-wordwide.org, 1996
Hill M, McColl W, Skillicorn D (1997) Questions and answers about BSP. Scientific Programming 6(3):249–274
Kielmann T, Bal HE, Verstoep K (2000) Fast measurement of LogP parameters for message passing platforms. In: 4th Workshop on Runtime Systems for Parallel Programming (RTSPP), held in conjunction with IPDPS 2000, May 2000
Kielmann T, Bal HE, Gorlatch S, Verstoep K, Hofman RFH (2001) Network performance-aware collective communication for clustered wide area systems. Parallel Computing 27(11): 1431–1456
Martin RP, Culler DE (1999) NFS sensitivity to high performance networks. In: ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems, pp 71–82, 1999
Pješivac-Grbović J, Angskun T, Bosilca G, Fagg GE, Gabriel E, Dongarra JJ (2004) Performance analysis of MPI collective operations. In: 4th International Workshop on Performance Modeling, Evaluation, and Optimization of Parallel and Distributed Systems (PMEO-PDS’05), April 2004
Valiant LG (1990) A bridging model for parallel computation. Communications of the ACM 33(8):103–111
Rauber T, Rünger G (2010) Parallel programming: for multicore and cluster systems. Springer, New York
Hwang K, Xu Z (1998) Scalable parallel computing. WCB/McGraw-Hill, New York
Culler DE, Singh JP, Gupta A (1999) Parallel computer architecture – a hardware/software approach. Morgan Kaufmann, San Francisco
Gibbons PB, Matias Y, Ramachandran V (1999) Can a shared-memory model serve as a bridging-model for parallel computation? Theor Comput Syst 32(3):327–359
Aggarwal A, Chandra AK, Snir M (1990) Communication complexity of PRAMs. Theor Comput Sci 71:3–28
De la Torre P, Kruskal CP (1996) Submachine locality in the bulk synchronous setting. In: Proceedings of the EUROPAR’96, LNCS 1124, pp 352–358, Springer, Berlin, August 1996
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer Science+Business Media, LLC
About this entry
Cite this entry
Kielmann, T., Gorlatch, S. (2011). Bandwidth-Latency Models (BSP, LogP). In: Padua, D. (eds) Encyclopedia of Parallel Computing. Springer, Boston, MA. https://doi.org/10.1007/978-0-387-09766-4_189
Download citation
DOI: https://doi.org/10.1007/978-0-387-09766-4_189
Publisher Name: Springer, Boston, MA
Print ISBN: 978-0-387-09765-7
Online ISBN: 978-0-387-09766-4
eBook Packages: Computer ScienceReference Module Computer Science and Engineering