Utility-Based Cache Partitioning: A Low-Overhead, High-Performance, Runtime Mechanism to Partition Shared Caches

By Moinuddin K. Qureshi, Yale N. Patt. Micro, 2006

1,094 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/333da42f4369dff8ca905ad21ea6ee2f5dc99d55

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

MASK: Redesigning the GPU Memory Hierarchy to Support Multi-Application Concurrency (2018, background): “[74] propose TAP, a TLP-aware cache management mechanism that modifies the utility-based cache partitioning policy [109] and the baseline cache insertion policy [56] to lower cache space interference between GPGPU and CPU applications.”
Gaining insights into multicore cache partitioning: Bridging the gap between simulation and real systems (2008, methodology): “Meanwhile, a number of cache partitioning methods have been proposed with different optimization objectives, including performance [17] [11] [5] [2], fairness [8] [2] [12], and QoS (Quality of Service) [6] [10] [12].”
Q-clouds: managing performance interference effects for QoS-aware clouds (2010, positive): “For example, hardware techniques have been designed that dynamically partition cache resources based upon utility metrics [27], or integrate novel insertion policies to pseudopartition caches [36].”
PIPP: promotion/insertion pseudo-partitioning of multi-core shared caches (2009, methodology): “It is clear that while this approach does not incur any additional storage overhead for shadow tags, the contents of the marginal gain counters may be compromised, leading to inaccurate utility curves [36].”
Cache-Conscious Wavefront Scheduling (2012, background): “Other schemes attempt to manage interference among heterogeneous work-loads [35, 19] but each thread in our workload has roughly similar characteristics.”
Bubble-up: Increasing utilization in modern warehouse scale computers via sensible co-locations (2011, positive): “One direction that has attracted much research attention is the management of shared cache and bandwidth through techniques such as resource partitioning [4, 20, 27–34, 38], throttling [7] and adaptive cache replacement policies [14].”
Addressing shared resource contention in multicore processors via scheduling (2010, methodology): “SDC, and other solutions relying on stack distance profiles such as cache partitioning [ 17 , 22, 26], assumed that the dominant cause of performance degradation is contention for the space in the shared cache, i.e., when co-scheduled threads evict each other data from the shared cache.”
3D-Stacked Memory Architectures for Multi-core Processors (2008, background): “, IPC) but then continue simulating the program so that it continues to compete for the shared L2 cache and bus resources (this is effectively the same as some other recent multi-core simulation methodologies [35]).”
Tiered-latency DRAM: A low latency and low cost DRAM architecture (2013, background): “, [8, 17, 18, 36, 37, 38, 46]) have proposed sophisticated cache management policies in the context of processor SRAM caches.”
Heracles: Improving resource efficiency at scale (2015, background): “Most cache partitioning schemes have been evaluated with a utility-based policy that optimizes for aggregate throughput [64].”
Osmotic Computing: A New Paradigm for Edge/Cloud Integration (2016, methodology): “Multiple approaches have applied optimization [12] and performance measurement techniques [13]”
SCORE (2018, background): “Cache partitioning has been well studied in the CPU caches and storage systems [32, 49, 56, 65].”
Cache Telepathy: Leveraging Shared Resource Attacks to Learn DNN Architectures (2018, background): “Finally, one can add security features to performance-oriented cache partitioning proposals [49],[50],[51],[52], to achieve better trade-offs between security and performance.”
Cooperative Caching for Chip Multiprocessors (2006, methodology): “Qureshi and Patt [123] proposed UMON sampling mechanism to provide more precise measurement, and lookahead partitioning algorithm to handle workloads with non-convex miss rate curves.”