Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley

Yildiz, Orcun; Zhou, Amelie Chi; Ibrahim, Shadi

doi:10.1016/j.future.2018.03.029

Title: Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley

Abstract

Here, burst buffer is an effective solution for reducing the data transfer time and the I/O interference in HPC systems. Extending Burst Buffers (BBs) to handle Big Data applications is challenging because BBs must account for the large data inputs of Big Data applications and the Quality-of-Service (QoS) of HPC applications which are considered as first-class citizens in HPC systems. Existing BBs focus on only intermediate data of Big Data applications and incur a high performance degradation of both Big Data and HPC applications. We present Eley, a burst buffer solution that helps to accelerate the performance of Big Data applications while guaranteeing the QoS of HPC applications. To achieve this goal, Eley embraces interference-aware prefetching technique that makes reading data input faster while introducing low interference for HPC applications. Evaluations using a wide range of Big Data and HPC applications demonstrate that Eley improves the performance of Big Data applications by up to 30% compared to existing BBs while maintaining the QoS of HPC applications.

Authors:

Yildiz, Orcun ^[1]; Zhou, Amelie Chi ^[2]; Ibrahim, Shadi ^[3]

Univ. Rennes, Rennes (France); Argonne National Lab. (ANL), Lemont, IL (United States)
Shenzhen Univ., Shenzhen (China)
Inria, Nantes (France)

Publication Date:: Thu Mar 22 00:00:00 EDT 2018

Research Org.:: Argonne National Laboratory (ANL), Argonne, IL (United States)

Sponsoring Org.:: National Science Foundation (NSF); USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR); Shenzhen University

OSTI Identifier:: 1467628

Alternate Identifier(s):: OSTI ID: 1583001

Grant/Contract Number:: AC02-06CH11357

Resource Type:: Accepted Manuscript

Journal Name:: Future Generations Computer Systems

Additional Journal Information:: Journal Volume: 86; Journal Issue: C; Journal ID: ISSN 0167-739X

Publisher:: Elsevier

Country of Publication:: United States

Language:: English

Subject:: 97 MATHEMATICS AND COMPUTING; Big Data; Burst Buffers; HPC; Interference; MapReduce; Parallel File Systems; Prefetch

Citation Formats


                    Yildiz, Orcun, Zhou, Amelie Chi, and Ibrahim, Shadi. Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley.  United States: N. p., 2018. 
Web.  doi:10.1016/j.future.2018.03.029.

Copy to clipboard


                    Yildiz, Orcun, Zhou, Amelie Chi, & Ibrahim, Shadi. Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley.  United States.  https://doi.org/10.1016/j.future.2018.03.029

Copy to clipboard


                    Yildiz, Orcun, Zhou, Amelie Chi, and Ibrahim, Shadi. Thu .  
"Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley".  United States.  https://doi.org/10.1016/j.future.2018.03.029.  https://www.osti.gov/servlets/purl/1467628.

Copy to clipboard


                    
@article{osti_1467628,

  title        = {Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley},

  author       = {Yildiz, Orcun and Zhou, Amelie Chi and Ibrahim, Shadi},

  abstractNote = {Here, burst buffer is an effective solution for reducing the data transfer time and the I/O interference in HPC systems. Extending Burst Buffers (BBs) to handle Big Data applications is challenging because BBs must account for the large data inputs of Big Data applications and the Quality-of-Service (QoS) of HPC applications which are considered as first-class citizens in HPC systems. Existing BBs focus on only intermediate data of Big Data applications and incur a high performance degradation of both Big Data and HPC applications. We present Eley, a burst buffer solution that helps to accelerate the performance of Big Data applications while guaranteeing the QoS of HPC applications. To achieve this goal, Eley embraces interference-aware prefetching technique that makes reading data input faster while introducing low interference for HPC applications. Evaluations using a wide range of Big Data and HPC applications demonstrate that Eley improves the performance of Big Data applications by up to 30% compared to existing BBs while maintaining the QoS of HPC applications.},

  doi          = {10.1016/j.future.2018.03.029},

  journal      = {Future Generations Computer Systems},

  number       = C,

  volume       = 86,

  place        = {United States},

  year         = {Thu Mar 22 00:00:00 EDT 2018},

  month        = {Thu Mar 22 00:00:00 EDT 2018}

}

Copy to clipboard

Journal Article:

Free Publicly Available Full Text

Accepted Manuscript (Publisher)

Accepted Manuscript (DOE)

Publisher's Version of Record

https://doi.org/10.1016/j.future.2018.03.029

Other availability

Search WorldCat to find libraries that may hold this journal

Citation Metrics:

Cited by: 4 works

Citation information provided by
Web of Science

Save / Share:

Export Metadata

Save to My Library

Works referenced in this record:

MapReduce: simplified data processing on large clusters
journal, January 2008

Dean, Jeffrey; Ghemawat, Sanjay; Mehta, Brijesh
Communications of the ACM, Vol. 51, Issue 1
DOI: 10.1145/1327452.1327492

On the Root Causes of Cross-Application I/O Interference in HPC Storage Systems
conference, May 2016

Yildiz, Orcun; Dorier, Matthieu; Ibrahim, Shadi
2016 IEEE International Parallel and Distributed Processing Symposium (IPDPS)
DOI: 10.1109/IPDPS.2016.50

CALCioM: Mitigating I/O Interference in HPC Systems through Cross-Application Coordination
conference, May 2014

Dorier, Matthieu; Antoniu, Gabriel; Ross, Rob
2014 IEEE International Parallel & Distributed Processing Symposium (IPDPS), 2014 IEEE 28th International Parallel and Distributed Processing Symposium
DOI: 10.1109/IPDPS.2014.27

I/O-Aware Batch Scheduling for Petascale Computing Systems
conference, September 2015

Zhou, Zhou; Yang, Xu; Zhao, Dongfang
2015 IEEE International Conference on Cluster Computing (CLUSTER)
DOI: 10.1109/CLUSTER.2015.45

Damaris: Addressing Performance Variability in Data Management for Post-Petascale Simulations
journal, October 2016

Dorier, Matthieu; Antoniu, Gabriel; Cappello, Franck
ACM Transactions on Parallel Computing, Vol. 3, Issue 3
DOI: 10.1145/2987371

Performance Modelling and Analysis of Software-Defined Networking under Bursty Multimedia Traffic
journal, December 2016

Miao, Wang; Min, Geyong; Wu, Yulei
ACM Transactions on Multimedia Computing, Communications, and Applications, Vol. 12, Issue 5s
DOI: 10.1145/2983637

Enabling fast failure recovery in shared Hadoop clusters: Towards failure-aware scheduling
journal, September 2017

Yildiz, Orcun; Ibrahim, Shadi; Antoniu, Gabriel
Future Generation Computer Systems, Vol. 74
DOI: 10.1016/j.future.2016.02.015

Works referencing / citing this record:

Approaches of enhancing interoperations among high performance computing and big data analytics via augmentation
journal, August 2019

Pathak, Ajeet Ram; Pandey, Manjusha; Rautaray, Siddharth S.
Cluster Computing, Vol. 23, Issue 2
DOI: 10.1007/s10586-019-02960-y

Similar Records in DOE PAGES and OSTI.GOV collections:

Evaluating Burst Buffer Placement in HPC Systems

Conference Khetawat, Harsh ; Zimmer, Christopher ; Mueller, Frank ; ...

Burst buffers (BBs) are increasingly exploited in contemporary supercomputers to bridge the performance gap between compute and storage systems. The design of BBs, particularly the placement of these devices and the underlying network topology, impacts both performance and cost. As the cost of other components such as memory and accelerators is increasing, it is becoming more important that HPC centers provision BBs tailored to their workloads.This work contributes a provisioning system to provide accurate, multi-tenant simulations that model realistic application and storage workloads from HPC systems. The framework aids HPC centers in modeling their workloads against multiple network and BBmore »« less
Full Text Available
Design and Implementation of Burst Buffer Over-Subscription Scheme for HPC Storage Systems

Journal Article Bang, Jiwoo ; Sim, Alexander ; Lockwood, Glenn K. ; ... - IEEE Access

Burst Buffer is widely used in supercomputer centers to bridge the performance gap between computational power and the high-performance I/O systems. The primary role of Burst Buffer is to temporarily absorb the bursty I/O and reduce the heavy access on Parallel File System (PFS). However, the job resource manager on High-Performance Computer (HPC) systems prefers to use a dedicated Burst Buffer allocation approach, which eventually leads to the severely underutilized Burst Buffer resource. To improve the efficiency of using the expensive Burst Buffer resource, we analyze the I/O patterns on Burst Buffer in depth. We propose Burst Buffer over-subscription allocationmore »« less
https://doi.org/10.1109/ACCESS.2022.3233829
Toward Managing HPC Burst Buffers Effectively: Draining Strategy to Regulate Bursty I/O Behavior

Conference Tang, Kun ; Huang, Ping ; He, Xubin ; ...

HPC (high-performance computing) applications usually show bursty I/O behaviors. In order to expedite the applications, permanent storage systems are usually provisioned to serve such I/O bursts. Approaching the era of exascale computing, non-volatile RAM is introduced as burst buffers, to absorb the bursty bulk data and relax the I/O provisioning requirement of the permanent storage systems. However, without judiciously draining the burst buffers, I/O bursts are passed down to the underlying storage systems, which causes severe I/O contention issues.In order to minimize the I/O provisioning requirement and resolve the issues caused by I/O bursts, we propose a proactive draining schememore »« less
https://doi.org/10.1109/MASCOTS.2017.35
BBOS: Efficient HPC Storage Management via Burst Buffer Over-Subscription

Conference Sung, H ; Bang, J ; Kim, C ; ...

To avoid access to PFS, dedicated BB allocation is preferred despite of severe BB underutilization. Recently, new all-flash HPC storage systems with integrated BB and PFS are proposed, which speed up access to PFS. For this reason, we adopt BB over-subscription allocation method by allowing HPC applications to use BB only for I/O phase for improving BB utilization. Unfortunately, BB over-subscription aggravates I/O interference and demotion overhead from BB to PFS, resulting in degraded performance. To minimize the performance degradation, we develop an I/O scheduler to prevent I/O congestion and a new transparent data management system based on checkpoint/restart characteristicsmore »« less
https://doi.org/10.1109/CCGrid49817.2020.00-79

Full Text Available
An empirical study of I/O separation for burst buffers in HPC systems

Journal Article Koo, Donghun ; Lee, Jaehwan ; Liu, Jialin ; ... - Journal of Parallel and Distributed Computing

To meet the exascale I/O requirements for the High-Performance Computing (HPC), a new I/O subsystem, Burst Buffer, based on solid state drives (SSD), has been developed. However, the diverse HPC workloads and the bursty I/O pattern cause severe data fragmentation that requires costly garbage collection (GC) and increases the number of bytes written to the SSD. To address this data fragmentation challenge, a new multi-stream feature has been developed for SSDs. In this work, we develop an I/O Separation scheme called BIOS to leverage this multi-stream feature to group the I/O streams based on the user IDs. We propose amore »« less
https://doi.org/10.1016/j.jpdc.2020.10.007

Full Text Available

Similar Records

Title: Improving the Effectiveness of Burst Buffers for Big Data Processing in HPC Systems with Eley

Abstract

Citation Formats

MapReduce: simplified data processing on large clusters journal, January 2008

On the Root Causes of Cross-Application I/O Interference in HPC Storage Systems conference, May 2016

CALCioM: Mitigating I/O Interference in HPC Systems through Cross-Application Coordination conference, May 2014

I/O-Aware Batch Scheduling for Petascale Computing Systems conference, September 2015

Damaris: Addressing Performance Variability in Data Management for Post-Petascale Simulations journal, October 2016

Performance Modelling and Analysis of Software-Defined Networking under Bursty Multimedia Traffic journal, December 2016

Enabling fast failure recovery in shared Hadoop clusters: Towards failure-aware scheduling journal, September 2017

Approaches of enhancing interoperations among high performance computing and big data analytics via augmentation journal, August 2019

MapReduce: simplified data processing on large clusters
journal, January 2008

On the Root Causes of Cross-Application I/O Interference in HPC Storage Systems
conference, May 2016

CALCioM: Mitigating I/O Interference in HPC Systems through Cross-Application Coordination
conference, May 2014

I/O-Aware Batch Scheduling for Petascale Computing Systems
conference, September 2015

Damaris: Addressing Performance Variability in Data Management for Post-Petascale Simulations
journal, October 2016

Performance Modelling and Analysis of Software-Defined Networking under Bursty Multimedia Traffic
journal, December 2016

Enabling fast failure recovery in shared Hadoop clusters: Towards failure-aware scheduling
journal, September 2017

Approaches of enhancing interoperations among high performance computing and big data analytics via augmentation
journal, August 2019