System-Level Performance Metrics for Multiprogram Workloads
By Stijn Eyerman, Lieven Eeckhout. IEEE Micro, 2008
457 citations recorded by the source corpus.
Subfields: Computer Networks and Communications, Hardware and Architecture, Information Systems
View Paper: https://www.semanticscholar.org/paper/68073f621072d793e95b9562bf9a9245415d5a96
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
- Onur Mutlu
- Yoongu Kim
- Saugata Ghose
- Rachata Ausavarungnirun
- Vivek Seshadri
- Mor Harchol-Balter
- Donghyuk Lee
- Adwait Jog
In Hardware and Architecture
- Onur Mutlu
- Yoongu Kim
- Rachata Ausavarungnirun
- Saugata Ghose
- Vivek Seshadri
- Mor Harchol-Balter
- Donghyuk Lee
- Thomas Moscibroda
In Information Systems
- Saugata Ghose
- Yixin Luo
- Yoongu Kim
- Rachata Ausavarungnirun
- Mor Harchol-Balter
- Onur Mutlu
- Christopher J. Rossbach
- Thomas Moscibroda
Citations in the corpus
- Fairness via source throttling: a configurable and high-performance fairness substrate for multi-core memory systems (2010, unknown): “These metrics are commonly used in measuring multiprogram performance in computer architecture [Eyerman and Eeckhout 2008].”
- ATLAS: A scalable and high-performance scheduling algorithm for multiple memory controllers (2010, methodology): “We evaluate the fairness of ATLAS using two separate metrics: maximum slowdown and harmonic speedup [13].”
- MASK: Redesigning the GPU Memory Hierarchy to Support Multi-Application Concurrency (2018, methodology): “As a result, MASK greatly increases the average number of threads that can be scheduled during longlatency stalls, which in turn improves system throughput (weighted speedup [42, 43]) by 57.”
- The evicted-address filter: A unified mechanism to address both cache pollution and thrashing (2012, methodology): “We evaluate system performance using the weighted speedup metric [50], which is shown to correlate with the system throughput [11].”
- A case for bufferless routing in on-chip networks (2009, methodology): “We also measure thread turnaround time using Hmean-Speedup [13], which balances fairness and throughput [33], but since measured hmean- and weighted-speedup results are qualitatively very similar in all our measurements, we mainly present weighted speedups for brevity.”
- RowClone: Fast and energy-efficient in-DRAM bulk data copy and initialization (2013, methodology): “We evaluate multi-core runs using the weighted speedup metric, a widely-used measure of system throughput for multi-programmed workloads [14], as well as five other performance/fairness/bandwidth/energy metrics, as shown in Table 7.”