Parallelism Always Helps

Mak, Louis

doi:10.1137/s0097539794265402

Cited by 10 publications

(3 citation statements)

References 23 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, there is evidence that large amounts of parallelism is available for almost any kind of problems if a strong enough model of computation (e.g. PRAM) is implemented [24] [22] [36]. This leaves a seamless support for the legacy code the most important motivation for including the full NUMA support for future ESM CMPs.…”

Section: Support Full Numamentioning

confidence: 99%

A PRAM-NUMA Model of Computation for Addressing Low-TLP Workloads

Forsell¹

2011

IJNC

View full text Add to dashboard Cite

It is possible to implement the parallel random access machine (PRAM) on a chip multiprocessor (CMP) efficiently with an emulated shared memory (ESM) architecture to gain easy parallel programmability crucial to wider penetration of CMPs to general purpose computing. This implementation relies on exploitation of the slack of parallel applications to hide the latency of the memory system instead of caches, sufficient bisection bandwidth to guarantee high throughput, and hashing to avoid hot spots in intercommunication. Unfortunately this solution can not handle workloads with low thread-level parallelism (TLP) efficiently because then there is not enough parallel slackness available for hiding the latency. In this paper we show that integrating non-uniform memory access (NUMA) support to the PRAM implementation architecture can solve this problem and provide a natural way for migration of the legacy code written for a sequential or multi-core NUMA machine. The obtained PRAM-NUMA hybrid model is defined and architectural implementation of it is outlined on our ECLIPSE ESM CMP framework. A high-level programming language example is given.

show abstract

Section: Support Full Numamentioning

confidence: 99%

A PRAM-NUMA Model of Computation for Addressing Low-TLP Workloads

Forsell¹

2011

IJNC

View full text Add to dashboard Cite

show abstract

“…For example, the NC = P question asks if polynomial time computations can be captured by (log n) O(1) depth, bounded fan-in, polynomial size circuits. Over the years, numerous parallel simulations have been given for speeding up general (serial) computational models [13,14,3,9,10,8], each of which can be interpreted as a Boolean circuit family, where the number of processors in the parallel algorithm is polynomially related to the circuit size of the family. However, no known parallel simulation achieves an ω(1) speedup (i.e.…”

Section: Introductionmentioning

confidence: 99%

“…The only result we could find directly addressing this question is from Mak [10] in 1997, who shows a time t RAM can be sped up to εt + n by a CREW PRAM of t O(1) processors, for ε > 0.…”

Section: Introductionmentioning

confidence: 99%

Parallelizing time with polynomial circuits

Williams

2005

Proceedings of the Seventeenth Annual ACM Symposium on Parallelism in Algorithms and Architectures

View full text Add to dashboard Cite

We study the relatively old problem of asymptotically reducing the runtime of serial computations with polynomial size Boolean circuits. To the best of our knowledge, no progress on this problem has been formally reported in the literature for general computational models, although we observe that early work of Chandra, Stockmeyer, and Vishkin implies the existence of non-uniform unbounded fan-in circuits of t O (1) size and O( t log log n ) depth, for time t Turing machines. We give an algorithmic size-depth tradeoff for parallelizing time t random access Turing machines, a model at least as powerful as logarithmic cost RAMs. Our parallel simulation yields logspace-uniform t O(1) size, O(t/ log t) depth Boolean circuits having semi-unbounded fan-in gates. In fact, for appropriate d, uniformOne corollary is that any log-cost time t RAM can be simulated by a log-cost CRCW PRAM using t O(1) processors and O(t/ log t) time. This is a major improvement over previous parallel speedups, which could only guarantee an Ω(log t) speedup with an exponential number of processors.

show abstract

The Refutation of Amdahl’s Law and Its Variants

Dévai

2017

Computational Science and Its Applications – ICCSA 2017

View full text Add to dashboard Cite

Parallelism Always Helps

Cited by 10 publications

References 23 publications

A PRAM-NUMA Model of Computation for Addressing Low-TLP Workloads

A PRAM-NUMA Model of Computation for Addressing Low-TLP Workloads

Parallelizing time with polynomial circuits

The Refutation of Amdahl’s Law and Its Variants

Contact Info

Product

Resources

About