Amdahl's Law in the Multicore Era

By Mark D. Hill, Michael R. Marty. Computer, 2008

1,426 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/3a02beb0ffb573c31a6be82565f62ba7abe98bc7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

Memory and Information Processing in Neuromorphic Systems (2015, background): “…system to carry out computation, and Section V will present an example of a neuromorphic processor that implements such mechanisms and that can be used to endow autonomous behaving systems with learning abilities for adapting to changes in the environment and interacting with it in real-time.”
ad-heap: an Efficient Heap Data Structure for Asymmetric Multicore Processors (2014, background): “Heterogeneous environments further enable pipeline parallelism to minimize serial bottleneck in Amdahl’s Law [17, 29, 14, 26].”
Exponential laws of computing growth (2016, background): “18 Amdahl’s Law would seem to limit the speedup to considerably less than the number of cores, because it assumes parallelism comes from remodeling a serial algorithm into a parallel algorithm.”
State-of-the-art in heterogeneous computing (2010, background): “The optimal type and composition of processors, however, will vary from one application to another [13,80].”
Scaling the bandwidth wall: challenges in and avenues for CMP scaling (2009, background): “[12] M. D. Hill and M. R. Marty.”
Learning scheduling algorithms for data processing clusters (2018, background): “In principle, this is a straightforward application of Amdahl’s Law [21], but given a mixed workload, the specific curves and sweet spots are difficult to predict.”
The AI gambit: leveraging artificial intelligence to combat climate change—opportunities, challenges, and recommendations (2021, background): “…while the computational requirements of neural networks had grown at a similar pace until the introduction of chips with multiple processor cores (Hill and Marty 2008; Thompson et al. 2020, 8) backpropagation, which are central components to contemporary DL research, had already been…”
Multi-core for mobile phones (2009, background): “Hence, a few relatively powerful cores complement the multi-core parallelism [35] [36], despite the lower area and power efficiency of these larger cores (Pollacks rule [34]).”
Accelerating critical section execution with asymmetric multi-core architectures (2009, background): “Critical sections are different from Amdahl s serial bottleneck: during the execution of a critical section, other threads that do not need to execute the same critical section can make progress.”
Classifying Compound Structures in Satellite Images: A Compressed Representation for Fast Queries (2015, background): “Due to Amdahl’s law [55], the computation can be easily parallelized on a stripe basis while maintaining a good paral-TABLE In summary, the proposed compound structure representation can be obtained at a pace of 440 s/1000 km 2 /CPU while occupying about 10% of the storage space of the input…”
Dark silicon and the end of multicore scaling (2011, methodology): “Previous studies on multicore performance modeling [8, 9, 10, 15, 20, 22, 28] use Pollack’s rule [6] to denote the tradeo ﬀ be-tween transistor count and performance.”
The multikernel: a new OS architecture for scalable multicore systems (2009, background): “…hardware the cache-coherence protocol between CPUs ensures that the OS can continue to safely assume a single shared memory, networking problems like routing and congestion are well-known concerns on large-scale multiprocessors, and are now issues in commodity intra-machine interconnects [18].”