Oguz Selvitopi

Oguz Selvitopi

Auto-generated profile, based on publicly available information

Oguz Selvitopi

Applied Mathematics and Computational Research Division, Lawrence Berkeley National Laboratory, Berkeley, CA, USA

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

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

Representative Publications

View All Publications
15 September 2020 in BMC Bioinformatics
Verified
Journal Article
Open access
PDF file
MAMuaaz G. Awan
Muaaz G. Awan
JDJack Deslippe
Jack Deslippe
ABAydin Buluc
Aydin Buluc
OSOguz Selvitopi
Oguz Selvitopi
SHSteven Hofmeyr
Steven Hofmeyr
KYKatherine Yelick
Katherine Yelick
Background: Bioinformatic workflows frequently make use of automated genome assembly and protein clustering tools. At the core of most of these tools, a significant portion of execution time is spent in determining optimal local alignment between two sequences. This task is performed with the Smith-Waterman algorithm, which is a dynamic programming based method. With the advent of modern sequencing technologies and increasing size of both genome and protein databases, a need for faster Smith-Waterman implementations has emerged. Multiple SIMD strategies for the Smith-Waterman algorithm are available for CPUs. However, with the move of HPC facilities towards accelerator based architectures, a need for an efficient GPU accelerated strategy has emerged. Existing GPU based strategies have either been optimized for a specific type of characters (Nucleotides or Amino Acids) or for only a handful of application use-cases.Results: In this paper, we present ADEPT, a new sequence alignment strategy for GPU architectures that is domain independent, supporting alignment of sequences from both genomes and proteins. Our proposed strategy uses GPU specific optimizations that do not rely on the nature of sequence. We demonstrate the feasibility of this strategy by implementing the Smith-Waterman algorithm and comparing it to similar CPU strategies as well as the fastest known GPU methods for each domain. ADEPT’s driver enables it to scale across multiple GPUs and allows easy integration into software pipelines which utilize large scale computational systems. We have shown that the ADEPT based Smith-Waterman algorithm demonstrates a peak performance of 360 GCUPS and 497 GCUPs for protein based and DNA based datasets respectively on a single GPU node (8 GPUs) of the Cori Supercomputer. Overall ADEPT shows 10x faster performance in a node-to-node comparison against a corresponding SIMD CPU implementation.Conclusions: ADEPT demonstrates a performance that is either comparable or better than existing GPU strategies. We demonstrated the efficacy of ADEPT in supporting existing bionformatics software pipelines by integrating ADEPT in MetaHipMer a high-performance denovo metagenome assembler and PASTIS a high-performance protein similarity graph construction pipeline. Our results show 10% and 30% boost of performance in MetaHipMer and PASTIS respectively.

1 July 2021 in IEEE Transactions on Parallel and Distributed Systems
Verified
Journal Article
Closed access
AAAriful Azad
Ariful Azad
OSOguz Selvitopi
Oguz Selvitopi
THTaufique Hussain
Taufique Hussain
JGJohn R. Gilbert
John R. Gilbert
ABAydın Buluç
Aydın Buluç
Combinatorial algorithms such as those that arise in graph analysis, modeling of discrete systems, bioinformatics, and chemistry, are often hard to parallelize. The Combinatorial BLAS library implements key computational primitives for rapid development of combinatorial algorithms in distributed-memory systems. During the decade since its first introduction, the Combinatorial BLAS library has evolved and expanded significantly. This article details many of the key technical features of Combinatorial BLAS version 2.0, such as communication avoidance, hierarchical parallelism via in-node multithreading, accelerator support via GPU kernels, generalized semiring support, implementations of key data structures and functions, and scalable distributed I/O operations for human-readable files. Our article also presents several rules of thumb for choosing the right data structures and functions in Combinatorial BLAS 2.0, under various common application scenarios.

1 November 2022
Verified
Conference Paper
Open access
NDNan Ding
Nan Ding
SWSamuel Williams
Samuel Williams
HNHai Ah Nam
Hai Ah Nam
TGTaylor Groves
Taylor Groves
MAMuaaz Gul Awan
Muaaz Gul Awan
NWNicholas Wright
Nicholas Wright
Tightly-coupled HPC systems have rigid memory allocation and can result in expensive memory resource underutilization. As novel memory and network technologies mature, disaggregated memory systems are becoming a promising solution for future HPC systems. It allows workloads to use the available memory of the entire system. In this paper, we propose a design framework to explore the disaggregated memory system design space. The framework incorporates memory capacity, network bandwidth, and local and remote memory access ratio, and provides an intuitive approach to guide machine configurations based on technology trends and workload characteristics. We apply our framework to analyze eleven workloads from five computational scenarios, including AI training, data analysis, genomics, protein, and traditional HPC. We demonstrate the ability of our methodology to understand the potential and pitfalls on a disaggregated memory system and motivate machine configurations. Our methodology shows that the 10 out of our 11 applications/workflows can leverage disaggregated memory without affecting performance.

12 August 2024
Verified
Conference Paper
Open access
UMUjjaini Mukhopadhyay
Ujjaini Mukhopadhyay
ATAlok Tripathy
Alok Tripathy
OSOguz Selvitopi
Oguz Selvitopi
KYKatherine Yelick
Katherine Yelick
ABAydin Buluc
Aydin Buluc
Graph Neural Networks (GNNs) are a computationally efficient method to learn embeddings and classifications on graph data. However, GNN training has low computational intensity, making communication costs the bottleneck for scalability. Sparse-matrix dense-matrix multiplication (SpMM) is the core computational operation in full-graph training of GNNs. Previous work parallelizing this operation focused on sparsity-oblivious algorithms, where matrix elements are communicated regardless of the sparsity pattern. This leads to a predictable communication pattern that can be overlapped with computation and enables the use of collective communication operations at the expense of wasting significant bandwidth by communicating unnecessary data. We develop sparsity-aware algorithms that tackle the communication bottlenecks in GNN training with three novel approaches. First, we communicate only the necessary matrix elements. Second, we utilize a graph partitioning model to reorder the matrix and drastically reduce the amount of communicated elements. Finally, we address the high load imbalance in communication with a tailored partitioning model, which minimizes both the total communication volume and the maximum sending volume. We further couple these sparsity-exploiting approaches with a communication-avoiding approach (1.5D parallel SpMM) in which submatrices are replicated to reduce communication. We explore the tradeoffs of these combined optimizations and show up to 14 × improvement on 256 GPUs and on some instances reducing communication to almost zero resulting in a communication-free parallel training relative to a popular GNN framework based on communication-oblivious SpMM.

28 March 2022
Verified
Conference Paper
Open access
SMSrđan Milaković
Srđan Milaković
OSOguz Selvitopi
Oguz Selvitopi
INIsrat Nisa
Israt Nisa
ZBZoran Budimlić
Zoran Budimlić
ABAydin Buluç
Aydin Buluç
Computing the product of two sparse matrices (SpGEMM) is a fundamental operation in various combinatorial and graph algorithms as well as various bioinformatics and data analytics applications for computing inner-product similarities. For an important class of algorithms, only a subset of the output entries are needed, and the resulting operation is known as Masked SpGEMM since a subset of the output entries is considered to be "masked out". In this work, we investigate various novel algorithms and data structures for this rather challenging and important computation, and provide guidelines on how to design a fast Masked-SpGEMM for shared-memory architectures.

17 November 2019
Verified
Conference Paper
Open access
OSOguz Selvitopi
Oguz Selvitopi
CACevdet Aykanat
Cevdet Aykanat
This work tackles the communication challenges posed by the latency-bound applications with irregular communication patterns, i.e., applications with high average and/or maximum message counts. We propose a novel algorithm for reorganizing a given set of irregular point-to-point messages with the objective of reducing total latency cost at the expense of increased volume. We organize processes into a virtual process topology inspired by the k-ary n-cube networks and regularize irregular messages by imposing regular communication pattern(s) onto them. Exploiting this process topology, we propose a flexible store-and-forward algorithm to control the trade-off between latency and volume. Our approach is able to reduce the communication time of sparse-matrix multiplication with latency-bound instances drastically: up to 22.6× for 16K processes on a 3D Torus network and up to 7.2× for 4K processes on a Dragonfly network, with its performance getting better with increasing number of processes.