Extreme scale computing: Modeling the impact of system noise in multicore clustered systems

Seelam, Seetharami; Fong, Liana; Tantawi, Asser N.; Lewars, John; Divirgilio, John; Gildea, K.

doi:10.1109/ipdps.2010.5470398

Cited by 23 publications

(13 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These results lend credence to the notion that this solution would also work for a much larger deployment, assuming system jitter can be minimized [31]. Specifically, it would be possible to expand such computational problems to a larger deployment in FutureGrid [17], Chameleon Cloud [22], or even the planned NSF Comet machine at SDSC, scheduled to provide up to 2 Petaflops of computational power.…”

Section: Discussionmentioning

confidence: 62%

Supporting High Performance Molecular Dynamics in Virtualized Clusters using IOMMU, SR-IOV, and GPUDirect

Younge

Walters

Crago

et al. 2015

Proceedings of the 11th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments

View full text Add to dashboard Cite

Cloud Infrastructure-as-a-Service paradigms have recently shown their utility for a vast array of computational problems, ranging from advanced web service architectures to high throughput computing. However, many scientific computing applications have been slow to adapt to virtualized cloud frameworks. This is due to performance impacts of virtualization technologies, coupled with the lack of advanced hardware support necessary for running many high performance scientific applications at scale.By using KVM virtual machines that leverage both Nvidia GPUs and InfiniBand, we show that molecular dynamics simulations with LAMMPS and HOOMD run at near-native speeds. This experiment also illustrates how virtualized environments can support the latest parallel computing paradigms, including both MPI+CUDA and new GPUDirect RDMA functionality. Specific findings show initial promise in scaling of such applications to larger production deployments targeting large scale computational workloads.

show abstract

Section: Discussionmentioning

confidence: 62%

Supporting High Performance Molecular Dynamics in Virtualized Clusters using IOMMU, SR-IOV, and GPUDirect

Younge

Walters

Crago

et al. 2015

Proceedings of the 11th ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments

View full text Add to dashboard Cite

show abstract

“…Shen [2010] provides a characterization of request behavior variations using server applications and finds that the inter-core resource sharing on multicore platforms obfuscates the request execution performance. Some studies [Constantinou et al 2005;Teng et al 2009;Becchi and Crowley 2006] investigated the impact of thread migrations and context-switches on application performance, and some other [Nataraj et al 2007;De et al 2007;Mann and Mittaly 2009;Seelam et al 2010;Beckman et al 2008] also explored the impact of OS noise on application performance. Verghese et al [1996] developed page migration and replication policies for ccNUMA systems and showed that migration and replication polices improved memory locality and reduced the overall stall time.…”

Section: Related Workmentioning

confidence: 99%

Thread Tranquilizer

Pusukuri

Gupta

Bhuyan

2012

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

To realize the performance potential of multicore systems, we must effectively manage the interactions between memory reference behavior and the operating system policies for thread scheduling and migration decisions. We observe that these interactions lead to significant variations in the performance of a given application, from one execution to the next, even when the program input remains unchanged and no other applications are being run on the system. Our experiments with multithreaded programs, including the TATP database application, SPECjbb2005, and a subset of PARSEC and SPEC OMP programs, on a 24-core Dell PowerEdge R905 server running OpenSolaris confirms the above observation. In this work we develop Thread Tranquilizer, an automatic technique for simultaneously reducing performance variation and improving performance by dynamically choosing appropriate memory allocation and process scheduling policies. Thread Tranquilizer uses simple utilities available on modern Operating Systems for monitoring cache misses and thread context-switches and then utilizes the collected information to dynamically select appropriate memory allocation and scheduling policies. In our experiments, Thread Tranquilizer yields up to 98% (average 68%) reduction in performance variation and up to 43% (average 15%) improvement in performance over default policies of OpenSolaris. We also demonstrate that Thread Tranquilizer simultaneously reduces performance variation and improves performance of the programs on Linux. Thread Tranquilizer is easy to use as it does not require any changes to the application source code or the OS kernel.

show abstract

“…There are several studies that provide an in-depth analysis of noise and its impact on large-scale systems [1]- [3], [17], [18]. Beckman et al discuss a benchmark for quantification of noise referred to as the selfish benchmark [19].…”

Section: Related Workmentioning

confidence: 99%

Weighted locality-sensitive scheduling for mitigating noise on multi-core clusters

Kale

Bhatelé

Gropp

2011

2011 18th International Conference on High Performance Computing

View full text Add to dashboard Cite

Recent studies have shown that operating system (OS) interference, popularly called OS noise can be a significant problem as we scale to a large number of processors. One solution for mitigating noise is to turn off certain OS services on the machine. However, this is typically infeasible because full-scale OS services may be required for some applications. Furthermore, it is not a choice that an end user can make. Thus, we need an application-level solution.Building upon previous work that demonstrated the utility of within-node light-weight load balancing, we discuss the technique of weighted micro-scheduling and provide insights based on experimentation for two different machines with very different noise signatures. Through careful enumeration of the search space of scheduler parameters, we allow our weighted micro-scheduler to be dynamic, adaptive and tunable for a specific application running on a specific architecture. By doing this, we show how we can enable running scientific applications efficiently on a very large number of processors, even in the presence of noise.

show abstract

Extreme scale computing: Modeling the impact of system noise in multicore clustered systems

Cited by 23 publications

References 11 publications

Supporting High Performance Molecular Dynamics in Virtualized Clusters using IOMMU, SR-IOV, and GPUDirect

Supporting High Performance Molecular Dynamics in Virtualized Clusters using IOMMU, SR-IOV, and GPUDirect

Thread Tranquilizer

Weighted locality-sensitive scheduling for mitigating noise on multi-core clusters

Contact Info

Product

Resources

About