Combining malleability and I/O control mechanisms to enhance the execution of multiple applications

Singh, David E.; Carretero, Jesús

doi:10.1016/j.jss.2018.11.006

Cited by 6 publications

(6 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…By comparing the exclusively-collected and the shared-collected metrics, the system is able to identify if there is performance degradation. Using this information, the scheduler can make decisions about the scheduling (Aupy et al, 2019; Carretero et al, 2020; Cascajo et al, 2019, 2021, 2022; Ekanadham et al, 1995; Sing David, 2019), for instance migrating one of them to avoid the interference, or evaluating if that interference is mitigated when the number of processes is increased or decreased for one of them.…”

Section: Building Synthetic Micro-benchmarks: Application Clonesmentioning

confidence: 99%

“…Efficient balancing of computational and I/O resources is one of the existing challenges in the HPC infrastructures (Panziera, 2017). This balance depends on different factors, such as the interaction between the hardware components, the software stack and the performed operations by the applications (Sing David, 2019). All of these factors may interfere with the I/O producing performance variations.…”

Section: Building Synthetic Micro-benchmarks: Application Clonesmentioning

confidence: 99%

See 1 more Smart Citation

Detecting interference between applications and improving the scheduling using malleable application clones

Cascajo,

Singh,

Carretero

2023

The International Journal of High Performance Computing Applica

Self Cite

View full text Add to dashboard Cite

This paper presents a novel feature for improving the scheduling process based on the performance prediction and the detection of CPU and I/O interference between applications. This feature consists of using malleable synthetic benchmarks – called clones – that reproduce the behaviour of applications executed in a cluster. These proxies can be used with two objectives: to build large and representative datasets that can be used to train the machine learning algorithms for modelling the platform workload, and to evaluate in advance if two executing applications can potentially produce contention hazards related to the shared use of the system resources like CPU, cache memory or I/O bandwidth. The proposed framework generates application clones based on generic-purpose performance information collected from monitoring. Unlike other works based on the use of micro-architectures or metrics obtained from simulators, in the approach presented in this work, the application clones generate similar behaviour without extracting data from the binaries, avoiding the necessity of managing code or data from the applications. One advantage of this approach is that they can be shared securely because they have not been generated using any piece of the original code. In addition, the proposed clones are malleable, so they also able to model the application behaviour under a different number of processes. In this work, we show how the use of clones contributes to improving the application scheduling by reducing the number of evaluations that is necessary to perform, and to detect performance degradation (interference) without the necessity of involving the execution of the actual applications. We evaluate the proposed clone generation approach on two sets of benchmarks (CPU and I/O oriented) and several applications. We also compare the performance obtained during the execution of the proxies and the applications to show their similarity. Finally, we include an evaluation of the interference detection using this novel approach.

show abstract

Section: Building Synthetic Micro-benchmarks: Application Clonesmentioning

confidence: 99%

Section: Building Synthetic Micro-benchmarks: Application Clonesmentioning

confidence: 99%

Detecting interference between applications and improving the scheduling using malleable application clones

Cascajo,

Singh,

Carretero

2023

The International Journal of High Performance Computing Applica

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the presence of competing applications, other solutions try to use the elasticity of applications and resources. Recently, Singh and Carretero Singh and Carretero (2019) proposed a middleware that would allow using the malleability of an application to shift forward or backward the time when an application is supposed to do its I/O movements. The solution AHPI/OS Isaila et al (2008) allows using elastic partitions that can scale up or down the number of storage resources available to an application.…”

Section: Architecture and I/o Optimizationmentioning

confidence: 99%

IO-aware Job-Scheduling: Exploiting the Impacts of Workload Characterizations to select the Mapping Strategy

Jeannot

Pallez

Vidal

2023

The International Journal of High Performance Computing Applica

View full text Add to dashboard Cite

In high performance, computing concurrent applications are sharing the same file system. However, the bandwidth which provides access to the storage is limited. Therefore, too many I/O operations performed at the same time lead to conflicts and performance loss due to contention. This scenario will become more common as applications become more data intensive. To avoid congestion, job-schedulers have to play an important role in selecting which application run concurrently. However I/O-aware mapping strategies need to be simple, robust and fast. Hence, in this article, we discuss two plain and practical strategies to mitigate I/O congestion. They are based on the idea of scheduling I/O access so as not to exceed some prescribed I/O bandwidth. More precisely, we compare two approaches: one grouping applications into packs that will be run independently (i.e., pack-scheduling), the other one scheduling greedily applications using a predefined order (i.e. list-scheduling). Results show that performances depend heavily on the I/O load and the homogeneity of the underlying workload. Finally, we introduce the notion of characteristic time that represents information on the average time between consecutive I/O transfers. We show that it could be important to the design of schedulers and that we expect it to be easily obtained by analysis tools.

show abstract

“…A second solution is to use the elasticity of applications or file system levels. Singh et al [25] present a strategy that combines I/O conflict prediction and application malleability. By means of prediction, the future I/O conflicts are forecast.…”

Section: Related Workmentioning

confidence: 99%

“…• We pick the proportion β of applications with low I/O load at random on [0, 1] (0 meaning all applications have high I/O load, 1 meaning they all have low I/O load). • We generate the number of applications of each workload uniformly at random in [25,100].…”

Section: Synthetic Workload Generationmentioning

confidence: 99%

Mapping and scheduling HPC applications for optimizing I/O

Carretero

Jeannot

Pallez

et al. 2020

Proceedings of the 34th ACM International Conference on Supercomputing

Self Cite

View full text Add to dashboard Cite

In HPC platforms, concurrent applications are sharing the same file system. This can lead to conflicts, especially as applications are more and more data intensive. I/O contention can represent a performance bottleneck. The access to bandwidth can be split in two complementary yet distinct problems. The mapping problem and the scheduling problem. The mapping problem consists in selecting the set of applications that are in competition for the I/O resource. The scheduling problem consists then, given I/O requests on the same resource, in determining the order to these accesses to minimize the I/O time. In this work we propose to couple a novel bandwidth-aware mapping algorithm to I/O list-scheduling policies to develop a cross-layer optimization solution.We study this solution experimentally using an I/O middleware: CLARISSE. We show that naive policies such as FIFO perform relatively well in order to schedule I/O movements, and that the important part to reduce congestion lies mostly on the mapping part. We evaluate the algorithm that we propose using a simulator that we validated experimentally. This evaluation shows important gains for the simple, bandwidth-aware mapping solution that we provide compared to its non bandwidth-aware counterpart. The gains are both in terms of machine efficiency (makespan) and application efficiency (stretch). This stresses even more the importance of designing efficient, bandwidth-aware mapping strategies to alleviate the cost of I/O congestion.

show abstract

Combining malleability and I/O control mechanisms to enhance the execution of multiple applications

Cited by 6 publications

References 48 publications

Detecting interference between applications and improving the scheduling using malleable application clones

Detecting interference between applications and improving the scheduling using malleable application clones

IO-aware Job-Scheduling: Exploiting the Impacts of Workload Characterizations to select the Mapping Strategy

Mapping and scheduling HPC applications for optimizing I/O

Contact Info

Product

Resources

About