Analysis of parallel algorithms

This article discusses the analysis of parallel algorithms. Like in the analysis of "ordinary", sequential, algorithms, one is typically interested in asymptotic bounds on the resource consumption (mainly time spent computing), but the analysis is performed in the presence of multiple processor units that cooperate to perform computations. Thus, one can determine not only how many "steps" a computation takes, but also how much faster it becomes as the number of processors goes up.

Overview

Suppose computations are executed on a machine that has $p$ processors. Let $T p$ denote the time that expires between the start of the computation and its end. Analysis of the computation's running time focuses on the following notions:

The work of a computation executed by $p$ processors is the total number of primitive operations that the processors perform.^[1] Ignoring communication overhead from synchronizing the processors, this is equal to the time used to run the computation on a single processor, denoted $T 1$ .
The span is the length of the longest series of operations that have to be performed sequentially due to data dependencies (the critical path). The span may also be called the critical path length or the depth of the computation.^[2] Minimizing the span is important in designing parallel algorithms, because the span determines the shortest possible execution time.^[3] Alternatively, the span can be defined as the time $T \infty$ spent computing using an idealized machine with an infinite number of processors.^[4]
The cost of the computation is the quantity $pT p$ . This expresses the total time spent, by all processors, in both computing and waiting.^[1]

Several useful results follow from the definitions of work, span and cost:

Work law. The cost is always at least the work: $pT p \geq T 1$ . This follows from the fact that $p$ processors can perform at most $p$ operations in parallel.^[1]^[4]
Span law. A finite number $p$ of processors cannot outperform an infinite number, so that $T p \geq T \infty$ .^[4]

Using these definitions and laws, the following measures of performance can be given:

Speedup is the gain in speed made by parallel execution compared to sequential execution: $S p = T 1 ∕ T p$ . When the speedup is $Ω(n)$ for input size $n$ (using big O notation), the speedup is linear, which is optimal in simple models of computation because the work law implies that $T 1 ∕ T p \leq p$ (super-linear speedup can occur in practice due to memory hierarchy effects). The situation $T 1 ∕ T p = p$ is called perfect linear speedup.^[4] An algorithm that exhibits linear speedup is said to be scalable.^[1]
Efficiency is the speedup per processor, $S p ∕ p$ .^[1]
Parallelism is the ratio $T 1 ∕ T \infty$ . It represents the maximum possible speedup on any number of processors. By the span law, the parallelism bounds the speedup: if $p > T 1 ∕ T \infty$ , then $T 1 ∕ T p \leq T 1 ∕ T \infty < p$ .^[4]
The slackness is $T 1 ∕ (pT \infty)$ . A slackness less than one implies (by the span law) that perfect linear speedup is impossible on $p$ processors.^[4]

Execution on a limited number of processors

Analysis of parallel algorithms is usually carried out under the assumption that an unbounded number of processors is available. This is unrealistic, but not a problem, since any computation that can run in parallel on $N$ processors can be executed on $p < N$ processors by letting each processor execute multiple units of work. A result called Brent's law states that one can perform such a "simulation" in time $T p$ , bounded by^[5]

T_{p}\leq T_{N}+{\frac {T_{1}-T_{N}}{p}},

or, less precisely,^[1]

T_{p}=O\left(T_{N}+{\frac {T_{1}}{p}}\right).

An alternative statement of the law bounds $T p$ above and below by

{\frac {T_{1}}{p}}\leq T_{p}<{\frac {T_{1}}{p}}+T_{\infty }

showing that the span (depth) $T \infty$ and the work $T 1$ together provide reasonable bounds on the computation time.^[2]

References

1 2 3 4 5 6 Casanova, Henri; Legrand, Arnaud; Robert, Yves (2008). Parallel Algorithms. CRC Press. p. 10.
1 2 Blelloch, Guy (1996). "Programming Parallel Algorithms" (PDF). Communications of the ACM. 39 (3): 85–97. doi:10.1145/227234.227246.
↑ Michael McCool; James Reinders; Arch Robison (2013). Structured Parallel Programming: Patterns for Efficient Computation. Elsevier. pp. 4–5.
1 2 3 4 5 6 Cormen, Thomas H.; Leiserson, Charles E.; Rivest, Ronald L.; Stein, Clifford (2009) [1990]. Introduction to Algorithms (3rd ed.). MIT Press and McGraw-Hill. pp. 779–784. ISBN 0-262-03384-4.
↑ Gustafson, John L. (2011). "Brent's Theorem". Encyclopedia of Parallel Computing. pp. 182–185.

Parallel computing

General	Distributed computing Cloud computing High-performance computing

Levels	Bit Instruction Task Data Memory Loop Pipeline

Multithreading	Temporal Simultaneous Preemptive Cooperative

Theory	PRAM model Analysis of parallel algorithms Amdahl's law Gustafson's law Cost efficiency Karp–Flatt metric Slowdown Speedup

Elements	Process Thread Fiber Instruction window

Coordination	Multiprocessing Memory coherency Cache coherency Cache invalidation Barrier Synchronization Application checkpointing

Programming	Stream Processing Dataflow programming Models Implicit parallelism Explicit parallelism Concurrency Non-blocking algorithm

Hardware	Flynn's taxonomy SISD SIMD MISD MIMD Dataflow architecture Pipelined processor Superscalar processor Vector processor Multiprocessor symmetric asymmetric Memory shared distributed distributed shared UMA NUMA COMA Massively parallel computer Computer cluster Grid computer

APIs	Ateji PX Boost.Thread Charm++ Cilk Coarray Fortran CUDA Dryad C++ AMP Global Arrays MPI OpenMP OpenCL OpenHMPP OpenACC TPL PLINQ PVM POSIX Threads RaftLib UPC TBB

Problems	Embarrassingly parallel Software lockout Scalability Race condition Deadlock Livelock Starvation Deterministic algorithm Parallel slowdown

Category: parallel computing Media related to parallel computing at Wikimedia Commons

This article is issued from Wikipedia - version of the 9/21/2015. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.