Search NASA⌕ Search

SEARCH · Search NASA

Results for “storage cache”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5

Adding Data Management Services to Parallel File Systems

The objective of this project, called DAMASC for “Data Management in Scientific Computing”, is to coalesce data management with parallel file system management to present a declarative interface to scientists for managing, querying, and analyzing extremely large data sets efficiently and predictably. Managing extremely large data sets is a key challenge of exascale computing. The overhead, energy, and cost of moving massive volumes of data demand designs where computation is close to storage. In current architectures, compute/analysis clusters access data in a physically separate parallel file system and largely leave it scientist to reduce data movement. Over the past decades the high-end computing community has adopted middleware with multiple layers of abstractions and specialized file formats such as NetCDF-4 and HDF5. These abstractions provide a limited set of high-level data processing functions, but have inherent functionality and performance limitations: middleware that provides access to the highly structured contents of scientific data files stored in the (unstructured) file systems can only optimize to the extent that file system interfaces permit; the highly structured formats of these files often impedes native file system performance optimizations. We are developing Damasc, an enhanced high-performance file system with native rich data management services. Damasc will enable efficient queries and updates over files stored in their native byte-stream format while retaining the inherent performance of file system data storage via declarative queries and updates over views of underlying files. Damasc has four key benefits for the development of data-intensive scientific code: (1) applications can use important data-management services, such as declarative queries, views, and provenance tracking, that are currently available only within database systems; (2) the use of these services becomes easier, as they are provided within a familiar file-based ecosystem; (3) common optimizations, e.g., indexing and caching, are readily supported across several file formats, avoiding effort duplication; and (4) performance improves significantly, as data processing is integrated more tightly with data storage. Our key contributions are: SciHadoop which explores changes to MapReduce assumption by taking advantage of semantics of structured data while preserving MapReduce’s failure and resource management; DataMods which extends common abstractions of parallel file systems so they become programmable such that they can be extended to natively support a variety of data models and can be hooked into emerging distributed runtimes such as Stanford’s Legion; and Miso which combines Hadoop and relational data warehousing to minimize time to insight, taking into account the overhead of ingesting data into data warehousing.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

A Temporal Locality-Aware Page-Mapped Flash Translation Layer

The poor performance of random writes has been a cause of major concern which needs to be addressed to better utilize the potential of flash in enterprise-scale environments. We examine one of the important causes of this poor performance: the design of the flash translation layer (FTL) which performs the virtual-to-physical address translations and hides the erase-before-write characteristics of flash. We propose a complete paradigm shift in the design of the core FTL engine from the existing techniques with our Demand-Based Flash Translation Layer (DFTL) which selectively caches page- level address mappings. Our experimental evaluation using FlashSim with realistic enterprise-scale workloads endorses the utility of DFTL in enterprise-scale storage systems by demonstrating: 1) improved performance, 2) reduced garbage collection overhead and 3) better overload behavior compared with hybrid FTL schemes which are the most popular implementation methods. For example, a predominantly random-write dominant I/O trace from an OLTP application running at a large financial institution shows a 78% improvement in average response time (due to a 3-fold reduction in operations of the garbage collector), compared with the hybrid FTL scheme. Even for the well-known read-dominant TPC-H benchmark, for which DFTL introduces additional overheads, we improve system response time by 56%. Moreover, interestingly, when write-back cache on DFTL-based SSD is enabled, DFTL even outperforms the page-based FTL scheme, improving their response time by 72% in Financial trace.

Address Mapping Table↗

Blackcomb: Hardware-Software Co-design for Non-Volatile Memory in Exascale Systems

Summary of technical results of Blackcomb Memory Devices We explored various different memory technologies (STTRAM, PCRAM, FeRAM, and ReRAM). The progress can be classified into three categories, below. Modeling and Tool Releases Various modeling tools have been developed over the last decade to help in the design of SRAM or DRAM-based memory hierarchies. To explore new design opportunities that NVM technologies can bring to the designers, we have developed similar high-level models for NVM, including PCRAMsim [Dong 2009], NVSim [Dong 2012], and NVMain [Poremba 2012]. NVSim is a circuit-level model for NVM performance, energy, and area estimation, which supports various NVM technologies, including STT-RAM, PCRAM, ReRAM, and legacy NAND Flash. NVSim is successfully validated against industrial NVM prototypes, and it is expected to help boost architecture-level NVM-related studies. On the other side, NVMain is a cycle accurate main memory simulator designed to simulate emerging nonvolatile memories at the architectural level. We have released these models as open source tools and provided contiguous support to them. We also proposed PS3-RAM, which is a fast, portable and scalable statistical STT-RAM reliability analysis model [Wen 2012]. Design Space Exploration and Optimization With the support of these models, we explore different device/circuit optimization techniques. For example, in [Niu 2012a] we studied the power reduction technique for the application of ECC scheme in ReRAM designs and proposed to use ECC code to relax the BER (Bit Error Rate) requirement of a single memory to improve the write energy consumption and latency for both 1T1R and cross-point ReRAM designs. In [Xu 2011], we proposed a methodology to design STT-RAM for different optimization goals such as read performance, write performance and write energy by leveraging the trade-off between write current and write time of MTJ. We also studied the tradeoffs in building a reliable crosspoint ReRAM array [Niu 2012b]. We have conducted an in depth analysis of the circuit and system level design implications of multi-layer cross-point Resistive RAM (MLCReRAM) from performance, power and reliability perspectives [Xu 2013]. The objective of this study is to understand the design trade-offs of this technology with respect to the MLC Phase Change Memory (MLCPCM).Our MLC ReRAM design at the circuit and system levels indicates that different resistance allocation schemes, programming strategies, peripheral designs, and material selections profoundly affect the area, latency, power, and reliability of MLC ReRAM. Based on this analysis, we conduct two case studies: first we compare MLC ReRAM design against MLC phase-change memory (PCM) and multi-layer cross-point ReRAM design, and point out why multi-level ReRAM is appealing; second we further explore the design space for MLC ReRAM. Architecture and Application We explored hybrid checkpointing using phase-change memory for future exascale systems [Dong 2011] and showed that the use of nonvolatile memory for local checkpointing significantly increases the number of faults covered by local checkpoints and reduces the probability of a global failure in the middle of a global checkpoint to less than 1%. We also proposed a technique called i2WAP to mitigate the write variations in NVM-based last-level cache for the improvement of the NVM lifetime [Wang 2013]. Our wear leveling technique attempts to work around the limitations of write endurance by arranging data access so that write operations can be distributed evenly across all the storage cells. During our intensive research on fault-tolerant NVM design, we found that ECC cannot effectively tolerate hard errors from limited write endurance and process imperfection. Therefore, we devised a novel Point and Discard (PAD) architecture in in [ 2012] as a hard-error-tolerant architecture for ReRAM-based Last Level Caches. PAD improves the lifetime of ReRAM caches by 1.6X-440X under different process variations without performance overhead in the system's early life. We have investigated the applicability of NVM for persistent memory design [Zhao 2013]. New byte addressable NVM enables fast persistent memory that allows in-memory persistent data objects to be updated with much higher throughput. Despite the significant improvement, the performance of these designs is only 50% of the native system with no persistence support, due to the logging or copy-on-write mechanisms used to update the persistent memory. A challenge in this approach is therefore how to efficiently enable atomic, consistent, and durable updates to ensure data persistence that survives application and/or system failures. We have designed a persistent memory system, called Klin, that can provide performance as close as that of the native system. The Klin design adopts a non-volatile cache and a non-volatile main memory for constructing a multi-versioned durable memory system, enabling atomic updates without logging or copy-on-write. Our evaluation shows that the proposed Kiln mechanism can achieve up to 2X of performance improvement to NVRAM-based persistent memory employing write-ahead logging. In addition, our design has numerous practical advantages: a simple and intuitive abstract interface, microarchitecture-level optimizations, fast recovery from failures, and no redundant writes to slow non-volatile storage media. The work was published in MICRO 2013 and received Best Paper Honorable Mentioned Award.

97 MATHEMATICS AND COMPUTING↗

On-the-Fly Decompression and Rendering of Multiresolution Terrain

We present a streaming geometry compression codec for multiresolution, uniformly-gridded, triangular terrain patches that supports very fast decompression. Our method is based on linear prediction and residual coding for lossless compression of the full-resolution data. As simplified patches on coarser levels in the hierarchy already incur some data loss, we optionally allow further quantization for more lossy compression. The quantization levels are adaptive on a per-patch basis, while still permitting seamless, adaptive tessellations of the terrain. Our geometry compression on such a hierarchy achieves compression ratios of 3:1 to 12:1. Our scheme is not only suitable for fast decompression on the CPU, but also for parallel decoding on the GPU with peak throughput over 2 billion triangles per second. Each terrain patch is independently decompressed on the fly from a variable-rate bitstream by a GPU geometry program with no branches or conditionals. Thus we can store the geometry compressed on the GPU, reducing storage and bandwidth requirements throughout the system. In our rendering approach, only compressed bitstreams and the decoded height values in the view-dependent 'cut' are explicitly stored on the GPU. Normal vectors are computed in a streaming fashion, and remaining geometry and texture coordinates, as well as mesh connectivity, are shared and re-used for all patches. We demonstrate and evaluate our algorithms on a small prototype system in which all compressed geometry fits in the GPU memory and decompression occurs on the fly every rendering frame without any cache maintenance.

99 GENERAL AND MISCELLANEOUS↗

A TTL-based Approach for Content Placement in Edge Networks

Edge networks are promising to provide better services to users by provisioning computing and storage resources at the edge of networks. However, due to the uncertainty and diversity of user interests, content popularity, distributed network structure, cache sizes, it is challenging to decide where to place the content, and how long it should be cached. In this paper, we study the utility optimization of content placement at edge networks through timer-based (TTL) policies. We propose provably optimal distributed algorithms that operate at each network cache to maximize the overall network utility. Our TTL-based optimization model provides theoretical answers to how long each content must be cached, and where it should be placed in the edge network. Extensive evaluations show that our algorithm significantly outperforms path replication with conventional caching algorithms over some network topologies.

Panigrahy, Nitish K.↗

Single-node Partitioned-Memory for Huge Graph Analytics: Cost and Performance Trade-offs

Nonvolatile memory NVDIMMs, available as Intel Optane, are less expensive than DRAM and bring large byte-addressable storage within reach to many applications. Evaluations on graph analytics have shown promising performance only when DRAM is used as a hardware cache (Memory mode). An open question is whether graph applications can exploit Optane and DRAM directly (AppDirect mode) and achieving better-than-DRAM average bandwidth and run times. We evaluate Optane as a volatile pool on two large-scale graph applications with very different computational patterns, Grappolo and Ripples. We show that AppDirect mode can deliver better-than-DRAM performance, by allocating data structures to Optane and DRAM according to their access characteristics, resulting in higher average memory bandwidth and lower average latency. Memory mode provides DRAM-competitive performance with capacity equal to persistent memory. We demonstrate occasional 4x improvement using the latest AppDirect option and frequently observe competitive performance between Optane AppDirect Memory modes and DRAM.

Ghosh, Sayan↗

Large Scale Caching and Streaming of Training Data for Online Deep Learning

The training of deep neural network models on large data remains a difficult problem, despite progress towards scalable techniques. In particular, there is a mismatch between the random but predetermined order in which AI flows select training samples and the streaming I/O patterns for which traditional HPC data storage (e.g., parallel file systems) are designed. In addition, as more data are obtained, it is feasible neither simply to train learning models incrementally, due to catastrophic forgetting (i.e., bias towards new samples), nor to train frequently from scratch, due to prohibitive time and/or resource constraints. In this paper, we study data management techniques that combine caching and streaming with rehearsal support in order to enable efficient access to training samples in both offline training and continual learning. We revisit state-of-art streaming approaches based on data pipelines that transparently handle prefetching, caching, shuffling, and data augmentation, and discuss the challenges and opportunities that arise when combining these methods with data-parallel training techniques. We also report on preliminary experiments that evaluate the I/O overheads involved in accessing the training samples from a parallel file system (PFS) under several concurrency scenarios, highlighting the impact of the PFS on the design of the data pipelines.

data pipelines↗

VA EDH Advanced Software Pipeline Framework Report: Enhancing Automation and Scalability

The VA Environmental Determinants of Health (EDH) Advanced Software Pipeline Framework is designed to enhance the efficiency, scalability, and security of geospatial data processing workflows. This framework integrates modern data orchestration and containerization technologies, including Prefect for workflow automation, Docker for containerization, and PostgreSQL/PostGIS for geospatial data storage and analysis. It ensures standardized, reproducible, and automated data processing, supporting VA objectives related to substance use risk assessment and recovery research. The pipeline addresses key scalability and performance challenges through horizontal and vertical scaling, high-performance computing (HPC) integration, parallel processing, task caching, and dynamic resource allocation. These optimizations improve throughput and reduce latency, allowing the system to efficiently manage large and complex datasets. Additionally, security and compliance measures—such as data encryption (SSL), Role-Based Access Control (RBAC), and adherence to GDPR and HIPAA standards—safeguard sensitive information throughout data transmission and storage. A key implementation of this framework includes the automation of shelter list geolocation workflows, ensuring that up-to-date data is readily available for VA decision-making. Lessons learned from this project include the transition from in-memory processing to incremental storage writes, improving resource management and reliability. Future enhancements aim to expand automation, integrate AI-driven anomaly detection, and incorporate high-performance computing resources. This framework provides a scalable, secure, and adaptable solution for managing geospatial datasets, reinforcing the VA’s ability to support clinical and strategic initiatives through data-driven decision-making.

97 MATHEMATICS AND COMPUTING↗

Integration experiences and performance studies of A COTS parallel archive systems

Current and future Archive Storage Systems have been asked to (a) scale to very high bandwidths, (b) scale in metadata performance, (c) support policy-based hierarchical storage management capability, (d) scale in supporting changing needs of very large data sets, (e) support standard interface, and (f) utilize commercial-off-the-shelf(COTS) hardware. Parallel file systems have been asked to do the same thing but at one or more orders of magnitude faster in performance. Archive systems continue to move closer to file systems in their design due to the need for speed and bandwidth, especially metadata searching speeds such as more caching and less robust semantics. Currently the number of extreme highly scalable parallel archive solutions is very small especially those that will move a single large striped parallel disk file onto many tapes in parallel. We believe that a hybrid storage approach of using COTS components and innovative software technology can bring new capabilities into a production environment for the HPC community much faster than the approach of creating and maintaining a complete end-to-end unique parallel archive software solution. In this paper, we relay our experience of integrating a global parallel file system and a standard backup/archive product with a very small amount of additional code to provide a scalable, parallel archive. Our solution has a high degree of overlap with current parallel archive products including (a) doing parallel movement to/from tape for a single large parallel file, (b) hierarchical storage management, (c) ILM features, (d) high volume (non-single parallel file) archives for backup/archive/content management, and (e) leveraging all free file movement tools in Linux such as copy, move, ls, tar, etc. We have successfully applied our working COTS Parallel Archive System to the current world's first petaflop/s computing system, LANL's Roadrunner, and demonstrated its capability to address requirements of future archival storage systems.

99 GENERAL AND MISCELLANEOUS↗

Integration experiments and performance studies of a COTS parallel archive system

Current and future Archive Storage Systems have been asked to (a) scale to very high bandwidths, (b) scale in metadata performance, (c) support policy-based hierarchical storage management capability, (d) scale in supporting changing needs of very large data sets, (e) support standard interface, and (f) utilize commercial-off-the-shelf (COTS) hardware. Parallel file systems have been asked to do the same thing but at one or more orders of magnitude faster in performance. Archive systems continue to move closer to file systems in their design due to the need for speed and bandwidth, especially metadata searching speeds such as more caching and less robust semantics. Currently the number of extreme highly scalable parallel archive solutions is very small especially those that will move a single large striped parallel disk file onto many tapes in parallel. We believe that a hybrid storage approach of using COTS components and innovative software technology can bring new capabilities into a production environment for the HPC community much faster than the approach of creating and maintaining a complete end-to-end unique parallel archive software solution. In this paper, we relay our experience of integrating a global parallel file system and a standard backup/archive product with a very small amount of additional code to provide a scalable, parallel archive. Our solution has a high degree of overlap with current parallel archive products including (a) doing parallel movement to/from tape for a single large parallel file, (b) hierarchical storage management, (c) ILM features, (d) high volume (non-single parallel file) archives for backup/archive/content management, and (e) leveraging all free file movement tools in Linux such as copy, move, Is, tar, etc. We have successfully applied our working COTS Parallel Archive System to the current world's first petafiop/s computing system, LANL's Roadrunner machine, and demonstrated its capability to address requirements of future archival storage systems.

99 GENERAL AND MISCELLANEOUS↗

Optimizing raid performance with cache

We live in a world of increasingly complex applications and operating systems. Information is increasing at a mind-boggling rate. The consolidation of text, voice, and imaging represents an even greater challenge for our information systems. Which forced us to address three important questions: Where do we store all this information? How do we access it? And, how do we protect it against the threat of loss or damage? Introduced in the 1980s, RAID (Redundant Arrays of Independent Disks) represents a cost-effective solution to the needs of the information age. While fulfilling expectations for high storage, and reliability, RAID is sometimes subject to criticisms in the area of performance. However, there are design elements that can significantly enhance performance. They can be subdivided into two areas: (1) RAID levels or basic architecture. And, (2) enhancement schemes such as intelligent caching, support of tagged command queuing, and use of SCSI-2 Fast and Wide features.

Bouzari, Alex↗

CoNNeCT Baseband Processor Module

A document describes the CoNNeCT Baseband Processor Module (BPM) based on an updated processor, memory technology, and field-programmable gate arrays (FPGAs). The BPM was developed from a requirement to provide sufficient computing power and memory storage to conduct experiments for a Software Defined Radio (SDR) to be implemented. The flight SDR uses the AT697 SPARC processor with on-chip data and instruction cache. The non-volatile memory has been increased from a 20-Mbit EEPROM (electrically erasable programmable read only memory) to a 4-Gbit Flash, managed by the RTAX2000 Housekeeper, allowing more programs and FPGA bit-files to be stored. The volatile memory has been increased from a 20-Mbit SRAM (static random access memory) to a 1.25-Gbit SDRAM (synchronous dynamic random access memory), providing additional memory space for more complex operating systems and programs to be executed on the SPARC. All memory is EDAC (error detection and correction) protected, while the SPARC processor implements fault protection via TMR (triple modular redundancy) architecture. Further capability over prior BPM designs includes the addition of a second FPGA to implement features beyond the resources of a single FPGA. Both FPGAs are implemented with Xilinx Virtex-II and are interconnected by a 96-bit bus to facilitate data exchange. Dedicated 1.25- Gbit SDRAMs are wired to each Xilinx FPGA to accommodate high rate data buffering for SDR applications as well as independent SpaceWire interfaces. The RTAX2000 manages scrub and configuration of each Xilinx.

Yamamoto, Clifford K↗

Robotics Verification and Validation Strategies for Perseverance Rover Sampling and Caching

The Mars 2020 Sampling and Caching Subsystem(SCS) is the most complex robotic system ever fielded on a MarsRover. It includes a 5 degree-of-freedom Robotic Arm, coringdrill, gas Dust Removal Tool, interfaces for two turret-mountedinstruments, and an Adaptive Caching Assembly (ACA). TheACA is itself a complex robotic system, containing hardware tosupport docking and bit exchange, a 3 degree-of-freedomSample Handling Assembly for manipulating sample tubes,storage for several drill bits and sample tubes, and mechanismsto support observing and sealing samples collected by the drill.To successfully verify and validate the SCS hardware andsoftware and its integration with the Mars 2020 flight systemseveral key strategies were employed.The SCS Verification and Validation (V&V) program utilizedmultiple test venues with tiered levels of fidelity. These includedsimulation and visualization software environments, low fidelitydevelopment testbeds, testbeds with high fidelity SCS hardwareand commercial off-the shelf avionics, integrated systemtestbeds with flight-like avionics, and environmental testbedscapable of simulating Martian surface temperature andpressure. Multiple units of each SCS hardware componentmoved fluidly between test venues to accomplish myriadstandalone and coordinated test objectives. Test preparationand executions were performed by a diverse team of engineerswith training and technical ownership tailored for individualexperience and role. Despite significant differences between testvenues, the SCS V&V team established efficient and consistentprocesses and tools for procedure development, test execution,and data review that enabled personnel, as well as technicalproducts such as sequences and parameter configurations, toflow between venues effectively. A series of benchmark testsprovided evidence of performance consistency as elements weretransferred between venues and as system capability evolved.This paper provides an overview of the SCS V&V program andexplores several overarching strategies that enabled successfuloperation in the face of unprecedented complexity. Keyoutcomes of the SCS validation effort are summarized, alongwith lessons learned and beneficial integrations of validationtool and process innovations into Mars surface operations.

Brooks, Sawyer↗

Data Locality Enhancement of Dynamic Simulations for Exascale Computing (Final Report)

The development of modern processors exhibits two trends that complicate the optimizations of modern software. The first is the increasing sensitivity of processors' throughput to irregularities in computation. With more processors produced through a massive integration of simple cores, future systems will increasingly favor regular data-level parallel computations, but deviate from the needs of applications with complex patterns. Some evidences are already shown on Graphic Processing Units (GPU): Irregular data accesses (e.g., indirect references A[D[i]]) and conditional branches are limiting many GPU applications' performance at a level an order of magnitude lower than the peak of GPU. The second hardware trend is the growing gap between memory bandwidth and the aggregate speed—that is, the sum of all cores' computing power—of a Chip Multiprocessor (CMP). Despite the capped growth of the peak CPU speed, the aggregate speed of a CMP keeps increasing as more cores get into a single chip. It is expected that by 2018, node concurrency in an exascale system will increase by hundreds of times, whereas, memory bandwidth will expand by only 10 to 20 times. Consequently, data movement and storage is expected to consume more than 70% of the total system power. Bridging this gap is difficult; the complexities of modern CMP memory hierarchy make it even harder: Data cache becomes shared among computing units, and the sharing is often non-uniform—whether two computing units share a cache depends on their proximity and the level of the cache. On the recent IBM Power7 architecture, for instance, four hardware contexts (or SMT threads) in a core share the entire memory hierarchy, all cores in one chip share an on-chip L3 cache, and cores across chips share L3 and main memory through off-chip connections. These two trends complicate the translation of computing power into performance, especially for a program with either intensive data accesses or complex patterns in data accesses or control flow paths. Unfortunately, both attributes present and will persist in a class of important applications. For instance, many scientific simulations deal with a large volume of data. And meanwhile, as most real-world processes are non-uniform and evolving (e.g., the evolution of a galaxy or the process of a drug injection), both the computations and data accesses of these programs tend to be irregular and dynamically changing. Currently, the lack of support to these applications on modern CMP severely limits their performance. On GPU, as our recent study shows and other studies echo, performance enhancement of a factor of integers is possible when memory accesses or control flows are streamlined for a set of GPU applications. On multicore CPU, our studies show that traditional locality enhancement, for being oblivious to the new features of multicore memory hierarchy, may even cause large slowdown to data-intensive dynamic applications. The severity of the issues is expected to worsen as the two hardware trends continue. Some recent studies try to match software with the trends, but in a limited scope or manner. For irregularities on GPU, most studies focus on irregularities analyzable through static analysis (e.g., data accesses in regular loops). Dynamic irregularities are harder to address because the needed analysis and transformations typically have to happen at run time. Some other research resorts to hardware extensions, an actual adoption of which is unclear for the entailed space cost and complexity. For data locality, recent years have seen some exploitations of the new memory hierarchy on multicore for performance, but most of them are on process or thread scheduling, rather than program transformations. Our recent study reveals that program-level transformations may magnify the scheduling benefits by a factor of seven, concluding that program-level transformation should play a central role for data locality enhancement on modern CMP. But research in this direction has been sparse, and most have focused on data layout or cache performance modeling, rather than program transformations to match with the new memory hierarchy features. Overall, it is still an open question how to bridge the gap between dynamic computations and the two prominent properties of modern processors. The goal of this project is to develop a set of techniques and software tools to enhance the matching between memory accesses in dynamic simulations and the prominent features of modern and future CMP, alleviating the memory performance issues for petascale and exascale computing. This report summarizes the discoveries and products produced throughout this project. It includes free launch, a new software approach to overcoming the shortcomings of both methods; coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution; algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Supporting dynamic parallelism is important for GPU to benefit a broad range of applications. There are currently two fundamental ways for programs to exploit dynamic parallelism on GPU: a software-based approach with software-managed worklists, and a hardware-based approach through dynamic subkernel launches. Neither is satisfactory. The former is complicated to program and is often subject to some load imbalance; the latter suffers large runtime overhead. In this work, we propose free launch, a new software approach to overcoming the shortcomings of both methods. It allows programmers to use subkernel launches to express dynamic parallelism. It employs a novel compiler-based code transformation named subkernel launch removal to replace the subkernel launches with the reuse of parent threads. Coupled with an adaptive task assignment mechanism, the transformation reassigns the tasks in the subkernels to the parent threads with a good load balance. The technique requires no hardware extensions, immediately deployable on existing GPUs. It keeps the programming convenience of the subkernel launch-based approach while avoiding its large runtime overhead. Meanwhile, its superior load balancing makes it outperform manual worklist-based techniques by 3X on average. The work was published at the 48th Annual IEEE/ACM International Symposium on Microarchitecture (Micro'2015). A Graphic Processing Unit (GPU) system is typically equipped with many types of memory (e.g., global, constant, texture, shared, cache). Data placement determines what data are placed on which type of memory, essential for GPU memory performance. Prior optimizations of data placement always require a single view of a data object on memory, which limits the optimization effectiveness. In this work, we propose coherence-free multiview, an approach that allows multiple views of a single data object to co-exist on GPU memory during a GPU kernel execution. We demonstrate that under certain conditions, the multiple views can remain incoherent while facilitating enhanced data placement. We present a theorem and some compiler support to ensure the soundness of the usage of coherence-free multiview. We further develop reference-discerning data placement, a new way to enhance data placements on GPU. It enables more flexible data placements by using coherence-free multiview to leverage the slack in coherence requirement of some GPU programs. Experiments on three types of GPU systems show that, with less than 200KB space cost, the new data placement technique can provide a 1.6X average (up to 4.27X) speedup. In addition, we have examined algorithmic optimizations to data analytics problems, especially those that involve lots of distance calculations. Computing distances among data points is an essential part of many important algorithms in data analytics, graph analysis, and other domains. In each of these domains, developers have spent significant manual effort optimizing algorithms, often through novel applications of the triangle equality, in order to minimize the number of distance computations in the algorithms. In this work, we observe that many algorithms across these domains can be generalized as an instance of a generic distance-related abstraction. Based on this abstraction, we derive seven principles for correctly applying the triangular inequality to optimize distance-related algorithms. Guided by the findings, we develop Triangular OPtimizer (TOP), the first software framework that is able to automatically produce optimized algorithms that either matches or outperforms manually designed algorithms for solving distance-related problems. TOP achieves up to 237x speedups and 2.5X on average. The work has been published at the 32nd International Conference on Machine Learning (ICML'15) and the 41st International Conference on Very Large Data Bases (VLDB'15).

42 ENGINEERING↗

DENOVA: Deduplication Extended NOVA File System

This paper shows mathematically and experimentally that inline deduplication is not suitable for file systems on ultra-low latency Intel Optane DC PM devices in terms of performance, and proposes DeNova, an offline deduplication specially designed for log-structured NVM file systems such as NOVA. DeNova offers high-performance and low-latency I/O processing and executes deduplication in the background without interfering with foreground I/Os. DeNova employs DRAM-free persistent deduplication metadata, favoring CPU cache line, and ensures failure consistency on any system failure. We implement DeNova in the NOVA file system. Evaluation with DeNova confirms a negligible performance drop of baseline NOVA of less than 1%, while gaining high storage space savings. Extensive experiments show DeNova is failure consistent in all failure scenario cases.

Khan, Awais↗

Mars Ascent Vehicle Development Status

The Mars robotic sample return mission has been a potential flagship mission for NASA s science mission directorate for decades. The Mars Exploration Program and the planetary science decadal survey have highlighted both the science value of the Mars Sample Return (MSR) mission, but also the need for risk reduction through technology development. One of the critical elements of the MSR mission is the Mars Ascent Vehicle (MAV), which must launch the sample cache from the surface of Mars and place it into low Mars orbit. The MAV has significant challenges to overcome due to tight constraints on the MAV s mass and volume, as well as environmental challenges associated with long duration storage on the Martian surface and during Entry Descent and Landing (EDL). In the fall of 2010, NASA selected three industrial partners for study phase contracts to develop MAV system concepts, identify technology needs, and recommend technology developments plans for follow-on work. In addition to the contractor recommendations, JPL s Team-X was used for a comparative assessment of the three vehicle concepts to understand relative strengths, weaknesses, and sensitivity to system growth. The GRC COMPASS team independently evaluated MAV system solutions using liquid bipropellant, solid rocket motors, and an advanced monopropellant option. The results of the study phase contracts and comparative assessment is provided herein.

Dankanich, John W.↗

Active Measurement of Memory Resource Consumption

Hierarchical memory is a cornerstone of modern hardware design because it provides high memory performance and capacity at a low cost. However, the use of multiple levels of memory and complex cache management policies makes it very difficult to optimize the performance of applications running on hierarchical memories. As the number of compute cores per chip continues to rise faster than the total amount of available memory, applications will become increasingly starved for memory storage capacity and bandwidth, making the problem of performance optimization even more critical. We propose a new methodology for measuring and modeling the performance of hierarchical memories in terms on the application’s utilization of the key memory resources: capacity of a given memory level and bandwidth between two levels. This is done by actively interfering with the application’s use of these resources. The application’s sensitivity to reduced resource availability is measured by observing the effect of interference on application performance. The resulting resource-oriented model of performance both greatly simplifies application performance analysis and makes it possible to predict an application’s performance when running with various resource constraints. This is useful to predict performance for future memory-constrained architectures.

97 MATHEMATICS AND COMPUTING↗

A scheme for supporting distributed data structures on multicomputers

A data migration mechanism is proposed that allows an explicit and controlled mapping of data to memory. While read or write copies of each data element can be assigned to any processor's memory, longer term storage of each data element is assigned to a specific location in the memory of a particular processor. The proposed integration of a data migration scheme with a compiler is able to eliminate the migration of unneeded data that can occur in multiprocessor paging or caching. The overhead of adjudicating multiple concurrent writes to the same page or cache line is also eliminated. Data is presented that suggests that the scheme may be a pratical method for efficiently supporting data migration.

Hiranandani, Seema↗