Gokcen Kestor

Gokcen Kestor

Auto-generated profile, based on publicly available information

Gokcen Kestor

Barcelona Supercomputing Center, Barcelona, Spain

Research Focus & Metrics

Academic Career Length

16Years (2011-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 September 2013
Verified
Conference Paper
Closed access
GKGokcen Kestor
Gokcen Kestor
RGRoberto Gioiosa
Roberto Gioiosa
DKDarren J. Kerbyson
Darren J. Kerbyson
AHAdolfy Hoisie
Adolfy Hoisie
In the exascale era, the energy cost of moving data across the memory hierarchy is expected to be two orders of magnitude higher than the cost of performing a double-precision floating point operation. Despite its importance, the energy cost of data movement in scientific applications has not be quantitatively evaluated even for current systems.

15 November 2015
Verified
Conference Paper
Open access
PDF file
RARizwan A. Ashraf
Rizwan A. Ashraf
RGRoberto Gioiosa
Roberto Gioiosa
GKGokcen Kestor
Gokcen Kestor
RDRonald F. DeMara
Ronald F. DeMara
CCChen-Yong Cher
Chen-Yong Cher
PBPradip Bose
Pradip Bose
Resiliency of exascale systems has quickly become an important concern for the scientific community. Despite its importance, still much remains to be determined regarding how faults disseminate or at what rate do they impact HPC applications. The understanding of where and how fast faults propagate could lead to more efficient implementation of application-driven error detection and recovery. In this work, we propose a fault propagation framework to analyze how faults propagate in MPI applications and to understand their vulnerability to faults. We employ a combination of compiler-level code transformation and instrumentation, along with a runtime checker. Using the information provided by our framework, we employ machine learning technique to derive application fault propagation models that can be used to estimate the number of corrupted memory locations at runtime.

1 May 2017
Verified
Conference Paper
Closed access
IPIvy Bo Peng
Ivy Bo Peng
RGRoberto Gioiosa
Roberto Gioiosa
GKGokcen Kestor
Gokcen Kestor
PCPietro Cicotti
Pietro Cicotti
ELErwin Laure
Erwin Laure
SMStefano Markidis
Stefano Markidis
Hardware accelerators have become a de-facto standard to achieve high performance on current supercomputers and there are indications that this trend will increase in the future. Modern accelerators feature high-bandwidth memory next to the computing cores. For example, the Intel Knights Landing (KNL) processor is equipped with 16 GB of high-bandwidth memory (HBM) that works together with conventional DRAM memory. Theoretically, HBM can provide ~4× higher bandwidth than conventional DRAM. However, many factors impact the effective performance achieved by applications, including the application memory access pattern, the problem size, the threading level and the actual memory configuration. In this paper, we analyze the Intel KNL system and quantify the impact of the most important factors on the application performance by using a set of applications that are representative of scientific and data-analytics workloads. Our results show that applications with regular memory access benefit from MCDRAM, achieving up to 3× performance when compared to the performance obtained using only DRAM. On the contrary, applications with random memory access pattern are latency-bound and may suffer from performance degradation when using only MCDRAM. For those applications, the use of additional hardware threads may help hide latency and achieve higher aggregated bandwidth when using HBM.

18 June 2017
Verified
Conference Paper
Open access
IPIvy Bo Peng
Ivy Bo Peng
RGRoberto Gioiosa
Roberto Gioiosa
GKGokcen Kestor
Gokcen Kestor
PCPietro Cicotti
Pietro Cicotti
ELErwin Laure
Erwin Laure
SMStefano Markidis
Stefano Markidis
Traditional scientific and emerging data analytics applications require fast, power-efficient, large, and persistent memories. Combining all these characteristics within a single memory technology is expensive and hence future supercomputers will feature different memory technologies side-by-side. However, it is a complex task to program hybrid-memory systems and to identify the best object-to-memory mapping. We envision that programmers will probably resort to use default configurations that only require minimal interventions on the application code or system settings. In this work, we argue that intelligent, fine-grained data placement can achieve higher performance than default setups. We present an algorithm for data placement on hybrid-memory systems. Our algorithm is based on a set of single-object allocation rules and global data placement decisions. We also present RTHMS, a tool that implements our algorithm and provides recommendations about the object-to-memory mapping. Our experiments on a hybrid memory system, an Intel Knights Landing processor with DRAM and HBM, show that RTHMS is able to achieve higher performance than the default configuration. We believe that RTHMS will be a valuable tool for programmers working on complex hybrid-memory systems.

16 February 2022
Verified
Book Chapter
Closed access
PDF file
EMErdal Mutlu
Erdal Mutlu
RTRuiqin Tian
Ruiqin Tian
BRBin Ren
Bin Ren
SKSriram Krishnamoorthy
Sriram Krishnamoorthy
RGRoberto Gioiosa
Roberto Gioiosa
GKGokcen Kestor
Gokcen Kestor

1 May 2017
Verified
Conference Paper
Closed access
GKGokcen Kestor
Gokcen Kestor
SKSriram Krishnamoorthy
Sriram Krishnamoorthy
WMWenjing Ma
Wenjing Ma
Nested fork-join programs scheduled using work stealing can automatically balance load and adapt to changes in the execution environment. In this paper, we design an approach to efficiently recover from faults encountered by these programs. Specifically, we focus on localized recovery of the task space in the presence of fail-stop failures. We present an approach to efficiently track, under work stealing, the relationships between the work executed by various threads. This information is used to identify and schedule the tasks to be re-executed without interfering with normal task execution. The algorithm precisely computes the work lost, incurs minimal re-execution overhead, and can recover from an arbitrary number of failures. Experimental evaluation demonstrates low overheads in the absence of failures, recovery overheads on the same order as the lost work, and much lower recovery costs than alternative strategies.