Automatically characterizing large scale program behavior

By Timothy Sherwood, Erez Perelman, Greg Hamerly, Brad Calder. ASPLOS X, 2002

1,885 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture, Information Systems

View Paper: https://www.semanticscholar.org/paper/443d3fa6ae8069e4867bf98577b16de66b104cb5

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

In Information Systems

Citations in the corpus

Thirty Years of Machine Learning: The Road to Pareto-Optimal Wireless Networks (2019, methodology): “while their Manhattan distance is calculated as [202]:”
Razor: a low-power pipeline based on circuit-level timing speculation (2003, methodology): “We used the SimPoint toolset’s Early SimPoints to pinpoint program locations that were highly representative of the entire program execution [ 18 ].”
Phase tracking and prediction (2003, methodology): “We note that the hashing function we use is fundamentally the same as the random projection method we used to generate phases in [21].”
Predicting whole-program locality through reuse distance analysis (2003, background): “i) consisted of two identical phases with mainly four repeating patterns [36].”
Memory Performance Attacks: Denial of Memory Service in Multi-Core Systems (2007, methodology): “The portions of applications that are simulated are determined using the SimPoint tool [32], which selects simulation points in the application that are representative of the application’s behavior as a whole.”
Techniques for Multicore Thermal Management: Classification and New Exploration (2006, methodology): “Using SimPoint [31], we simulate representative traces of 500 million instructions from each program.”
Techniques to reduce the soft error rate of a high-performance microprocessor (2004, methodology): “We obtained the number of instructions to skip using Sherwood, et al.,’s [23] SimPoint analysis modified for the IA64 instruction set architecture.”
An Analysis of Efficient Multi-Core Global Power Management Policies: Maximizing Performance for a Given Power Budget (2006, background): “The different regions of execution are commonly referred to as phases of application behavior [39].”
Runtime power monitoring in high-end processors: methodology and empirical data (2003, methodology): “Several prior papers have proposed methods for detecting or exploiting program phases[1, 5, 9, 22, 23, 29].”
Benchmarking modern multiprocessors (2011, methodology): “To reduce the number of simulated instructions of a benchmark, a commonly used method is to take a subset of the instructions with statistical sampling, which requires a sophisticated mechanism in the simulator [17, 20, 80, 97].”
Reducing power density through activity migration (2003, methodology): “SimPoints [9] were used to fast-forward benchmark initialization phase and choose a representative 100 Million-Cycle section of each program.”
Live, Runtime Phase Monitoring and Prediction on Real Systems with Application to Dynamic Power Management (2006, background): “In recent years, studies have demonstrated different approaches to track application phase behavior, relying on execution properties such as control flow [7, 10, 15, 20, 22, 23] or power/performance characteristics [3, 5, 6, 12, 13, 26, 27].”
Sniper: Exploring the level of abstraction for scalable and accurate parallel multi-core simulation (2011, neutral): “These simulation points are chosen either randomly [7], periodically [31] or through phase analysis [26].”
Shoestring: probabilistic soft error reliability on the cheap (2010, positive): “Once an injection site is determined, program simulation is allowed to run in native mode (running on real hardware) until it reaches a representative code segment (identified using SimPoint [28] and manual source code inspection).”
McPAT: An integrated power, area, and timing modeling framework for multicore and manycore architectures (2009, positive): “We ﬁnd the representative simulation phases of each application and their weights using Simpoint 3.0 [39].”
System-Level Performance Metrics for Multiprogram Workloads (2008, methodology): “For example, the SimPoint tool selects fairly large simulation points containing on the order of millions of instructions to represent an entire program execution.(7) Alameldeen and Wood describe ways of defining units of work for multithreaded workloads, such as transactions completed.”
Efficiently exploring architectural design spaces via predictive modeling (2006, methodology): “[34] combine basic block distribution analysis with clustering to summarize behavior of sections of program execution; this guides selection of representative samples to simulate in detail.”