Seed location filtering is critical in DNA read mapping, a process where billions of DNA fragments (reads) sampled from a donor are mapped onto a reference genome to identify genomic variants of the donor. State-of-the-art read mappers 1) quickly generate possible mapping locations for seeds (i.e., smaller segments) within each read, 2) extract reference sequences at each of the mapping locations, and 3) check similarity between each read and its associated reference sequences with a computationally-expensive algorithm (i.e., sequence alignment) to determine the origin of the read. A seed location filter comes into play before alignment, discarding seed locations that alignment would deem a poor match. The ideal seed location filter would discard all poor match locations prior to alignment such that there is no wasted computation on unnecessary alignments. We propose a novel seed location filtering algorithm, GRIM-Filter, optimized to exploit 3D-stacked memory systems that integrate computation within a logic layer stacked under memory layers, to perform processing-in-memory (PIM). GRIM-Filter quickly filters seed locations by 1) introducing a new representation of coarse-grained segments of the reference genome, and 2) using massively-parallel in-memory operations to identify read presence within each coarse-grained segment. Our evaluations show that for a sequence alignment error tolerance of 0.05, GRIM-Filter 1) reduces the false negative rate of filtering by 5.59x–6.41x, and 2) provides an end-to-end read mapper speedup of 1.81x–3.65x, compared to a state-of-the-art read mapper employing the best previous seed location filtering algorithm. GRIM-Filter exploits 3D-stacked memory, which enables the efficient use of processing-in-memory, to overcome the memory bandwidth bottleneck in seed location filtering. We show that GRIM-Filter significantly improves the performance of a state-of-the-art read mapper. GRIM-Filter is a universal seed location filter that can be applied to any read mapper. We hope that our results provide inspiration for new works to design other bioinformatics algorithms that take advantage of emerging technologies and new processing paradigms, such as processing-in-memory using 3D-stacked memory devices.

/pdf/grim-filter-fast-seed-location-filtering-in-dna-read-mapping-1zabiexcp7.pdf

GRIM-Filter: Fast seed location filtering in DNA read mapping using processing-in-memory technologies.

A survey of optimization techniques for thermal-aware 3D processors

Computational genomics has proven its great potential to support precise and customized health care. However, with the wide adoption of the Next Generation Sequencing (NGS) technology, 'DNA Alignment', as the crucial step in computational genomics, is becoming more and more challenging due to the booming bio-data. Consequently, various hardware approaches have been explored to accelerate DNA seeding - the core and most time consuming step in DNA alignment. Most previous hardware approaches leverage multi-core, GPU, and FPGA to accelerate DNA seeding. However, DNA seeding is bounded by memory and above hardware approaches focus on computation. For this reason, Near Data Processing (NDP) is a better solution for DNA seeding. Unfortunately, existing NDP accelerators for DNA seeding face two grand challenges, i.e., fine-grained random memory access and scalability demand for booming bio-data. To address those challenges, we propose a practical, energy efficient, Dual-Inline Memory Module (DIMM) based, NDP Accelerator for DNA Seeding Algorithm (MEDAL), which is based on off-the-shelf DRAM components. For small databases that can be fitted within a single DRAM rank, we propose the intra-rank design, together with an algorithm-specific address mapping, bandwidth-aware data mapping, and Individual Chip Select (ICS) to address the challenge of fine-grained random memory access, improving parallelism and bandwidth utilization. Furthermore, to tackle the challenge of scalability for large databases, we propose three inter-rank designs (polling-based communication, interrupt-based communication, and Non-Volatile DIMM (NVDIMM)-based solution). In addition, we propose an algorithm-specific data compression technique to reduce memory footprint, introduce more space for the data mapping, and reduce the communication overhead. Experimental results show that for three proposed designs, on average, MEDAL can achieve 30.50x/8.37x/3.43x speedup and 289.91x/6.47x/2.89x energy reduction when compared with a 16-thread CPU baseline and two state-of-the-art NDP accelerators, respectively.

https://dl.acm.org/doi/pdf/10.1145/3352460.3358329

MEDAL: Scalable DIMM based Near Data Processing Accelerator for DNA Seeding Algorithm

Motivation: Seed location filtering is critical in DNA read mapping, a process where billions of DNA fragments (reads) sampled from a donor are mapped onto a reference genome to identify genomic variants of the donor. State-of-the-art read mappers 1) quickly generate possible mapping locations for seeds (i.e., smaller segments) within each read, 2) extract reference sequences at each of the mapping locations, and 3) check similarity between each read and its associated reference sequences with a computationally-expensive algorithm (i.e., sequence alignment) to determine the origin of the read. A seed location filter comes into play before alignment, discarding seed locations that alignment would deem a poor match. The ideal seed location filter would discard all poor match locations prior to alignment such that there is no wasted computation on unnecessary alignments. 
Results: We propose a novel seed location filtering algorithm, GRIM-Filter, optimized to exploit 3D-stacked memory systems that integrate computation within a logic layer stacked under memory layers, to perform processing-in-memory (PIM). GRIM-Filter quickly filters seed locations by 1) introducing a new representation of coarse-grained segments of the reference genome, and 2) using massively-parallel in-memory operations to identify read presence within each coarse-grained segment. Our evaluations show that for a sequence alignment error tolerance of 0.05, GRIM-Filter 1) reduces the false negative rate of filtering by 5.59x--6.41x, and 2) provides an end-to-end read mapper speedup of 1.81x--3.65x, compared to a state-of-the-art read mapper employing the best previous seed location filtering algorithm. 
Availability: The code is available online at: this https URL

/pdf/grim-filter-fast-seed-location-filtering-in-dna-read-mapping-1hema6wm8c.pdf

GRIM-Filter: Fast Seed Location Filtering in DNA Read Mapping Using Processing-in-Memory Technologies

Data movement between the CPU and main memory is a first-order obstacle against improv ing performance, scalability, and energy efficiency in modern systems. Computer systems employ a range of techniques to reduce overheads tied to data movement, spanning from traditional mechanisms (e.g., deep multi-level cache hierarch ies, aggressive hardware prefetcher s) to emerging techniques such as Near-Data Processing (NDP), where some computation is moved close to memory. Prior NDP works investigate the root causes of data movement bottlenecks using different profiling methodologies and tools. However, there is still a lack of understanding about the key metrics that can identify different data movement bottlenecks and their relation to traditional and emerging data movement mitigation mechanisms. Our goal is to methodically identify potential sources of data movement over a broad set of applications and to comprehensively compare traditional compute-centric data movement mitigation techniques (e.g., cach ing and prefetch ing) to more memory-centric techniques (e.g., NDP), thereby developing a rigorous understanding of the best techniques to mitigate each source of data movement. With this goal in mind, we perform the first large-scale characterization of a wide variety of applications, across a wide range of application domains, to identify fundamental program properties that lead to data movement to/from main memory. We develop the first systematic methodology to classify applications based on the sources contributing to data movement bottlenecks. From our large-scale characterization of 77K functions across 345 applications, we select 144 functions to form the first open-source benchmark suite (DAMOV) for main memory data movement studies. We select a diverse range of functions that (1) represent different types of data movement bottlenecks, and (2) come from a wide range of application domains. Using NDP as a case study, we identify new insights about the different data movement bottlenecks and use these insights to determine the most suitable data movement mitigation mechanism for a particular application. We open-source DAMOV and the complete source code for our new characterization methodology at https://github.com/CMU-SAFARI/DAMOV .

/pdf/damov-a-new-methodology-and-benchmark-suite-for-evaluating-48po45odv4.pdf

DAMOV: A New Methodology and Benchmark Suite for Evaluating Data Movement Bottlenecks

Sequence analysis plays extremely important role in bioinformatics, and most applications of which have compute intensive kernels consuming over 70% of total execution time. By exploiting the compute intensive execution stages of popular sequence analysis applications, we present and evaluate a VLSI architecture with a focus on those that target at biological sequences directly, including pairwise sequence alignment, multiple sequence alignment, database search, and short read sequence mappings. Based on coarse grained reconfigurable array we propose the use of many-core and 3D-stacked technologies to gain further improvement over memory subsystem, which gives another order of magnitude speedup from high bandwidth and low access latency. We analyze our approach in terms of its throughput and efficiency for different application mappings. Initial experimental results are evaluated from a stripped down implementation in a commodity FPGA, and then we scale the results to estimate the performance of our architecture with 9 layers of $$70 \hbox { mm}^{2}$$
 stacked wafers in 45-nm process. We demonstrate numerous estimated speedups better than corresponding existed hardware accelerator platforms for at least 40 times for the entire range of applications and datasets of interest. In comparison, the alternative FPGA based accelerators deliver only improvement for single application, while GPGPUs perform not well enough on accelerating program kernel with random memory access and integer addition/comparison operations.

/pdf/3d-stacked-many-core-architecture-for-biological-sequence-5czi77h8aa.pdf

3D-Stacked Many-Core Architecture for Biological Sequence Analysis Problems

Sequence analysis plays critical role in bioinformatics, and most applications of which have compute intensive kernels consuming over 70% of total execution time. By exploiting the compute intensive execution stages of popular sequence analysis applications, we present and evaluate a VLSI architecture with a focus on those that target at biological sequences directly, including pairwise alignment, multiple sequence alignment, database search, and short read sequence mappings. Based on coarse grained reconfigurable array (CGRA) we propose the use of many-core and 3D-stacked technologies to gain further improvement over memory subsystem, which gives another order of magnitude speedup from high bandwidth and low access latency. We analyze our approach in terms of its throughput and efficiency for different application mappings. Initial experimental results are evaluated from a stripped down implementation in a commodity FPGA, and then we scale the results to estimate the performance of our architecture with 9 layers of 68 mm2 stacked wafers in 45-nm process. We demonstrate numerous estimated speedups better than any existed hardware accelerators for at least 39 times for the entire range of applications and datasets of interest. In comparison, the alternative FPGA based accelerators deliver only improvement for single application, while GPGPUs perform not well enough on accelerating program kernel with random memory access and integer addition/comparison operations.

3D-stacked many-core architecture for biological sequence analysis problems

A coarse-grained reconfigurable processor tailored for accelerating multiple bioinformatics algorithms is proposed. In this paper, a programmable and scalable architectural platform instantiates an array of coarse grained light weight processing elements, which allows arbitrary partitioning, scheduling schemes and capable of solving complete four popular bioinformatics algorithms: the Needleman-Wunsch, Smith-Waterman, and HMMER on sequencing, and Maximum Likelihood on phylogenetic. The key difference of the proposed CGRA based solution compared to FPGA and GPU based solutions is a much better match on architecture and algorithms for the core computational needs, as well as the system level architectural needs. For the same degree of parallelism, we provide a 5X to 14X speed-up improvements compared to FPGA solutions and 15X to 78X compared to GPU acceleration on 3 sequencing algorithms. We also provide 2.8X speed-up compared to FPGA with the same amount of core logic and 70X compared to GPU with the same silicon area for Maximum Likelihood.

A Coarse-Grained Reconfigurable Processor for Sequencing and Phylogenetic Algorithms in Bioinformatics

A Coarse Grain Reconfigurable Architecture (CGRA) tailored for accelerating bio-informatics algorithms is proposed. The key innovation is a light weight bio-informatics processor that can be reconfigured to perform different Add Compare and Select operations of the popular sequencing algorithms. A programmable and scalable architectural platform instantiates an array of such processing elements and allows arbitrary partitioning and scheduling schemes and capable of solving complete sequencing algorithms including the sequential phases and deal with arbitrarily large sequences. The key difference of the proposed CGRA based solution compared to FPGA and GPU based solutions is a much better match of the architecture and algorithm for the core computational need as well as the system level architectural need. This claim is quantified for three popular sequencing algorithms: the Needleman-Wunsch, Smith-Waterman and HMMER. For the same degree of parallelism, we provide a 5 X and 15 X speed-up improvements compared to FPGA and GPU respectively. For the same size of silicon, the advantage grows by a factor of another 10 X.

A Coarse Grain Reconfigurable Architecture for sequence alignment problems in bio-informatics

Matching nucleic acid sequences has long become the performance bottleneck in genome assembly which aims to connect enormous partial genome reads without prior knowledge of the reference sequence. The intensive and random data accesses of querying sequences using the widely adopted FM-Index data structure have caused in-efficient usage and long runtime of the memory system. Existing software FM-Index tools are limited on algorithmic inefficiency and poor processing parallelism. Solutions on GPU, FPGA and ASIC focus mainly on computational acceleration while still bottlenecked at the memory-bound nature of querying FM-Index. This paper proposes DSIM, a scalable FM-Index querying on near-DRAM accelerators. DSIM supports highly parallel multi-step query processing by distributing partial FM-Index table to different DRAM chips. Each genome sequence is partitioned into shorter queries and dispatched to the corresponding DRAM chip for string lookup. The optimized data layout and execution control on DRAM enables high row-data reuse and minimizes CPU-DRAM data transfers. The light-weight mapping scheme on the host CPU facilitates effective query distribution to DRAM chips and further supports scalability to multiple DIMMs (Dual-Inline Memory Modules). An in-DRAM arbiter is implemented to control the intra-chip data processing without affecting the external memory controller and DDR protocol. Experiments on 128-chip DRAM system showed that DSIM achieves up to 231 $\times$ and 8.9 $\times$ overall speedup compared to the software FM-Index tool and the state-of-the-art near-DRAM solution respectively.

Pei Liu

Papers

3D-Stacked Many-Core Architecture for Biological Sequence Analysis Problems

3D-stacked many-core architecture for biological sequence analysis problems

A Coarse-Grained Reconfigurable Processor for Sequencing and Phylogenetic Algorithms in Bioinformatics

A Coarse Grain Reconfigurable Architecture for sequence alignment problems in bio-informatics

DSIM: Distributed Sequence Matching on Near-DRAM Accelerator for Genome Assembly