Exploring Data Migration for Future Deep-Memory Many-Core Systems

Perarnau, Swann; Zounmevo, Judicael A.; Gerofi, Balazs; Iskra, Kamil; Beckman, Pete

doi:10.1109/cluster.2016.42

Cited by 8 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A solution based on re-purposing arenabased heap management to keep locality among related data structures that are used together is discussed in [23]. Two mechanisms available on modern Linux systems for migrating data between physical locations are analysed in [24]. Memory migration mechanisms within the Linux kernel are far behind a simple user-space memory copy, and, with additional software degradation, a regular NUMA system can reasonably approach the bandwidth of a deep-memory architecture.…”

Section: B Migrationmentioning

confidence: 99%

Survey of Memory Management Techniques for HPC and Cloud Computing

Pupykina

Agosta

2019

IEEE Access

View full text Add to dashboard Cite

The emergence of new classes of HPC applications and usage models, such as real-time HPC and cloud HPC, coupled with the increasingly heterogeneous nature of HPC architectures, requires a renewed investigation of memory management solutions. Traditionally, memory is shared by an operating system using segmentation and paging techniques. At the same time, new classes of applications require Quality of Service (QoS) guarantees. As such, the typical practice of reserving a subset of the supercomputer to a single application becomes less attractive, leading to the exploration of cloud technologies. In this context, a viable scenario is that of multiple applications, with different QoS levels, coexisting on the same deeply heterogeneous HPC infrastructure and sharing resources. However, for this scenario to succeed in practice, resources, including memory, need to be allocated with a vision that includes both the application requirements and the current and future state of the overall system. In this survey, challenges of memory management in HPC and Cloud Computing, different memory management systems and optimisation techniques to increase memory utilisation are discussed in detail. INDEX TERMS Clouds, high performance computing, memory management, resource management. ANNA PUPYKINA received the higher education degree in data processing and control systems engineering from Samara State Aerospace University, Russia, in 2009. She is currently pursuing the Ph.D. degree with the Politecnico di Milano, under the supervision of Prof. G. Agosta. She was with Russian State Humanitarian University (RSUH), Togliatti, and also with the Department of Applied Mathematics and Informatics, Togliatti State University (TSU). Her main research interests include runtime management algorithms and programming model extensions for deeply heterogeneous systems. GIOVANNI AGOSTA received the Laurea degree in computer engineering and the Ph.D. degree in information technology from the Politecnico di Milano, Italy, in 2000 and 2004, respectively. He is currently a tenured Researcher. He has authored more than 80 articles in international journals, conferences, and workshops. His main research interest includes compiler technology. He was a recipient of four HiPEAC Awards and three Best Paper Awards. He has been involved in eight European funded projects, since 2010, holding workpackage and task leader responsibilities. He is a Full Member of the HiPEAC Network and the Organizer of two workshops colocated with the HiPEAC Conference.

show abstract

Section: B Migrationmentioning

confidence: 99%

Survey of Memory Management Techniques for HPC and Cloud Computing

Pupykina

Agosta

2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…This section will summarize some upcoming changes to the existing technologies but also the expected impact of more speculative technologies that require more research before they find their way into data centers. An important trend is the addition of wider buses and asynchronous protocols for data movement in the form of NVMe and also the support for highbandwidth memory (HBM) [74]. HBM requires architecture changes, which are not backwards compatible with older hardware, but will bring significant benefit to bandwidth-bound applications [73].…”

Section: Technologiesmentioning

confidence: 99%

Survey of Storage Systems for High-Performance Computing

Lüttgau

Kuhn

Duwe

et al. 2018

JSFI

View full text Add to dashboard Cite

In current supercomputers, storage is typically provided by parallel distributed file systems for hot data and tape archives for cold data. These file systems are often compatible with local file systems due to their use of the POSIX interface and semantics, which eases development and debugging because applications can easily run both on workstations and supercomputers. There is a wide variety of file systems to choose from, each tuned for different use cases and implementing different optimizations. However, the overall application performance is often held back by I/O bottlenecks due to insufficient performance of file systems or I/O libraries for highly parallel workloads. Performance problems are dealt with using novel storage hardware technologies as well as alternative I/O semantics and interfaces. These approaches have to be integrated into the storage stack seamlessly to make them convenient to use. Upcoming storage systems abandon the traditional POSIX interface and semantics in favor of alternative concepts such as object and key-value storage; moreover, they heavily rely on technologies such as NVM and burst buffers to improve performance. Additional tiers of storage hardware will increase the importance of hierarchical storage management. Many of these changes will be disruptive and require application developers to rethink their approaches to data management and I/O. A thorough understanding of today's storage infrastructures, including their strengths and weaknesses, is crucially important for designing and implementing scalable storage systems suitable for demands of exascale computing.

show abstract

“…We use this graph as input for the simulations and run the scheduling and mapping heuristics presented in Section 5. For the experiments, we extend the previous study developed for parallel stencil applications in [19] and provide a deep-memory implementation of the 1D Gauss-Seidel kernel for the KNL architecture. First, we copy tiles to migrate input and output data between slow and fast memory.…”

Section: Experimental Settingsmentioning

confidence: 99%

“…Data migration addresses the issue of moving data dynamically across memories during the execution of the application. Preliminary work [19] on this approach showcased that performance of a simple stencil benchmark can be improved by migration, using a scheme similar to out-of-core algorithms, when the computedensity of the application kernel is high enough to provide compute/migration overlapping. Closer to the focus of this paper, another study [6] discussed a runtime method to schedule tasks with data dependencies on a deep memory platform.…”

Section: Introductionmentioning

confidence: 99%

A Performance Model to Execute Workflows on High-Bandwidth-Memory Architectures

Benoît

Perarnau

Pottier

et al. 2018

Proceedings of the 47th International Conference on Parallel Processing

Self Cite

View full text Add to dashboard Cite

This work presents a realistic performance model to execute scientific workflows on high-bandwidth-memory architectures such as the Intel Knights Landing. We provide a detailed analysis of the execution time on such platforms, taking into account transfers from both fast and slow memory and their overlap with computations. We discuss several scheduling and mapping strategies: not only tasks must be assigned to computing resources, but also one has to decide which fraction of input and output data will reside in fast memory and which will have to stay in slow memory. We use extensive simulations to assess the impact of the mapping strategies on performance. We also conduct experiments for a simple 1D Gauss-Seidel kernel, which assess the accuracy of the model and further demonstrate the importance of a tuned memory management. Our model and results lay the foundations for further studies and experiments on dual-memory systems.

show abstract

Exploring Data Migration for Future Deep-Memory Many-Core Systems

Cited by 8 publications

References 21 publications

Survey of Memory Management Techniques for HPC and Cloud Computing

Survey of Memory Management Techniques for HPC and Cloud Computing

Survey of Storage Systems for High-Performance Computing

A Performance Model to Execute Workflows on High-Bandwidth-Memory Architectures

Contact Info

Product

Resources

About