Characterizing multi-threaded applications based on shared-resource contention

Dey, Tanima; Wang, Wei; Davidson, Jack W.; Soffa, Mary Lou

doi:10.1109/ispass.2011.5762717

Cited by 32 publications

(35 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This observation is a major reason for workload colocation, a technique that increases energy efficiency by trading it against certain other aspects of performance that can suffer due to resource sharing [2][3][4]. Colocated workloads are very common-in fact, the mere use of multiprocessing technically qualifies as workload colocation.…”

Section: Introductionmentioning

confidence: 99%

Analyzing the Impact of CPU Pinning and Partial CPU Loads on Performance and Energy Efficiency

Podzimek

Bulej

Chen

et al. 2015

2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

Abstract-While workload colocation is a necessity to increase energy efficiency of contemporary multicore hardware, it also increases the risk of performance anomalies due to workload interference. Pinning certain workloads to a subset of CPUs is a simple approach to increasing workload isolation, but its effect depends on workload type and system architecture. Apart from common sense guidelines, the effect of pinning has not been extensively studied so far. In this paper we study the impact of CPU pinning on performance interference and energy efficiency for pairs of colocated workloads. Besides various combinations of workloads, virtualization and resource isolation, we explore the effects of pinning depending on the level of background load. The presented results are based on more than 1000 experiments carried out on an Intel-based NUMA system, with all power management features enabled to reflect real-world settings. We find that less common CPU pinning configurations improve energy efficiency at partial background loads, indicating that systems hosting colocated workloads could benefit from dynamic CPU pinning based on CPU load and workload type.

show abstract

Section: Introductionmentioning

confidence: 99%

Analyzing the Impact of CPU Pinning and Partial CPU Loads on Performance and Energy Efficiency

Podzimek

Bulej

Chen

et al. 2015

2015 15th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing

View full text Add to dashboard Cite

show abstract

“…Our simulator models a superscalar microprocessor similar to Alpha 21264 [5]. The important parameters of the processor and the memory hierarchy The cache vulnerability factor (CVF) is the AVF for cache memory resources.…”

Section: Evaluation Methodologymentioning

confidence: 99%

“…Because of contention, especially for the resources in the memory hierarchy, an application's performance can degrade by more than 50% [4], and scalable performance improvement is often not achieved on multicore and many-core machines [5]. Contention for the shared resources in the memory hierarchy can also lead to ine cient resource usage [3,6].…”

Section: Resource Contentionmentioning

confidence: 99%

“…co-running threads, as shown in Figure 1.1(b), and contend for more than one cache with a multi-threaded co-runner. Moreover, multi-threaded applications can have contentious behavior among its own threads even when they do not have any co-runner [5], as shown in Figure 1.1(c). In addition, single-threaded applications do not have data sharing, whereas, multi-threaded applications can have data sharing [11].…”

Section: Resource Contentionmentioning

confidence: 99%

“…Assume both applications have negative SensitivityScores performance of ≠a P and ≠b P , respectively and contentious behavior for shared L2-cache on the quad-core platform (e.g., Figure 4.2(a). These applications, with negative sensitivity scores, su er from relatively higher intra-application contention than inter-application contention for cache resources [5], and the performance degrades more when the sibling threads are co-located with each other compared to when threads are co-located with the threads from the co-running application to use the same cache. Therefore, for such workloads with negative SensitivityScores performance , it is beneficial to share the resource with the co-runner's threads than sharing the resource with its sibling threads to improve the workload's overall response time and throughput.…”

Section: Scenariomentioning

confidence: 99%

See 2 more Smart Citations

ReSense: A Unified Framework for Improving Performance and Reliability in Multicore Architectures

Dey

Self Cite

View full text Add to dashboard Cite

Chip-multiprocessors (CMPs) have become ubiquitous in modern computing and the mainstream architecture for various platforms, including laptops, desktops, and large server machines. As technology scaling continues and more transistors are accommodated on the chip, the number of cores on CMPs is growing, and multi-core machines are scaling up to many-core machines. With this multi-core scaling, two major problems arise: shared-resource contention and soft errors or transient faults. Shared-resource contention can degrade an application's performance significantly, and soft errors increase the probability of incorrect application execution and the production of visible errors. To realize the full potential of multi-and many-core platforms, it is critical to ensure that applications in a workload not only execute e ciently and fast, but also correctly.In this dissertation, we develop a novel, general, and unified framework, ReSense, to address several challenges on multicore architectures including performance optimization, reliability improvement, power and thermal management. The framework includes five components: a general characterization methodology, a characterization metric, a sensitivity score, a thread mapping algorithm, and a run-time system. An instance of the framework is applied in two phases: characterization and mapping. The characterization phase utilizes the general characterization methodology and characterization metric to identify application characteristics without considering any co-runner(s). It generates a resource-sensitivity score for each application in a workload. In the mapping phase, the run-time system uses a thread-mapping algorithm and the sensitivity scores of the applications in a workload to c Abstract d determine the thread-mappings that optimize the objective function of the targeted problem.To demonstrate the utility and e ectiveness of ReSense, we use it to address the problems of shared-resource contention and soft errors for multi-threaded applications. For the resource contention problem, the characterization methodology determines how a multi-threaded application's performance is a ected as it shares a resource in the memory hierarchy. A sensitivity score based on resource contention is produced for each application in a workload.The run-time system uses the resource-contention sensitivity scores and a thread-mapping algorithm to allocate threads from a workload to core to mitigate shared-resource contention, thus improving response time and throughput.For the soft error problem, the characterization methodology determines how a multithreaded application's vulnerability to soft errors in shared caches is a ected by its resource occupancy duration. A sensitivity score based on cache occupancy is produced for each application in a workload. The run-time system uses the cache-occupancy sensitivity scores and a thread-mapping algorithm to allocate workload threads to cores to reduce the occupancy in the shared caches, thus reducing cache vulnerability.Both minimizing...

show abstract

SymS: a symmetrical scheduler to improve multi‐threaded program performance on NUMA systems

Zhu

Jin

Liao

2015

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThe nonuniform memory access (NUMA) architecture has been used extensively in data centers. Most of the previous works used single-threaded multiprogrammed workloads to study the performance of NUMA systems, which mainly focus on two classes of problems: resource contention and data locality. However, when running multi-threaded programs on NUMA systems, the critical thread of these programs significantly influences the system performance and brings new challenges that are different from those in a single-threaded situation. In particular, an additional scheduling scheme is desired to avoid the performance degradation caused by the critical thread of multi-threaded programs running on NUMA systems. This work presents a scheduler, Symmetrical Scheduler, which successfully solves the lagging problem by balancing the number of the costly remote shared data accesses for threads on NUMA systems. To the best of our knowledge, little work has been conducted to examine the performance impacted by the critical thread of multi-threaded programs on NUMA systems. By running the PARSEC benchmark on such systems, our methodology can improve the program performance by a factor of 6% on average and achieve maximally 25.3% improvement compared with Linux kernel scheduling mechanism.

show abstract

Characterizing multi-threaded applications based on shared-resource contention

Cited by 32 publications

References 25 publications

Analyzing the Impact of CPU Pinning and Partial CPU Loads on Performance and Energy Efficiency

Analyzing the Impact of CPU Pinning and Partial CPU Loads on Performance and Energy Efficiency

ReSense: A Unified Framework for Improving Performance and Reliability in Multicore Architectures

SymS: a symmetrical scheduler to improve multi‐threaded program performance on NUMA systems

Contact Info

Product

Resources

About