2014 IEEE International Symposium on Workload Characterization (IISWC)最新文献

筛选
英文 中文
Characterization and analysis of dynamic parallelism in unstructured GPU applications 非结构化GPU应用中动态并行性的表征与分析
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983039
Jin Wang, S. Yalamanchili
{"title":"Characterization and analysis of dynamic parallelism in unstructured GPU applications","authors":"Jin Wang, S. Yalamanchili","doi":"10.1109/IISWC.2014.6983039","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983039","url":null,"abstract":"GPUs have been proven very effective for structured applications. However, emerging data intensive applications are increasingly unstructured - irregular in their memory and control flow behavior over massive data sets. While the irregularity in these applications can result in poor workload balance among fine-grained threads or coarse-grained blocks, one can still observe dynamically formed pockets of structured data parallelism that can locally effectively exploit the GPU compute and memory bandwidth. In this study, we seek to characterize such dynamically formed parallelism and and evaluate implementations designed to exploit them using CUDA Dynamic Parallelism (CDP) - an execution model where parallel workload are launched dynamically from within kernels when pockets of structured parallelism are detected. We characterize and evaluate such implementations by analyzing the impact on control and memory behavior measurements on commodity hardware. In particular, the study targets a comprehensive understanding of the overhead of current CDP support in GPUs in terms of kernel launch, memory footprint and algorithm overhead. Experiments show that while the CDP implementation can generate potentially 1.13x-2.73x speedup over non-CDP implementations, the non-trivial overhead causes the overall performance an average of 1.21x slowdown.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131061231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 81
Characterizing the performance effect of trials and rotations in applications that use Quantum Phase Estimation 描述在使用量子相位估计的应用中试验和旋转对性能的影响
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983057
S. Patil, Ali JavadiAbhari, Chen-Fu Chiang, Jeff Heckey, M. Martonosi, F. Chong
{"title":"Characterizing the performance effect of trials and rotations in applications that use Quantum Phase Estimation","authors":"S. Patil, Ali JavadiAbhari, Chen-Fu Chiang, Jeff Heckey, M. Martonosi, F. Chong","doi":"10.1109/IISWC.2014.6983057","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983057","url":null,"abstract":"Quantum Phase Estimation (QPE) is one of the key techniques used in quantum computation to design quantum algorithms which can be exponentially faster than classical algorithms. Intuitively, QPE allows quantum algorithms to find the hidden structure in certain kinds of problems. In particular, Shor's well-known algorithm for factoring the product of two primes uses QPE. Simulation algorithms, such as Ground State Estimation (GSE) for quantum chemistry, also use QPE. Unfortunately, QPE can be computationally expensive, either requiring many trials of the computation (repetitions) or many small rotation operations on quantum bits. Selecting an efficient QPE approach requires detailed characterizations of the tradeoffs and overheads of these options. In this paper, we explore three different algorithms that trade off trials versus rotations. We perform a detailed characterization of their behavior on two important quantum algorithms (Shor's and GSE). We also develop an analytical model that characterizes the behavior of a range of algorithms in this tradeoff space.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125602170","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 9
Characterization of call-graph profiles in Java workloads Java工作负载中调用图配置文件的特性描述
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983055
T. Nakaike, H. Inoue, T. Suganuma, Moriyoshi Ohara
{"title":"Characterization of call-graph profiles in Java workloads","authors":"T. Nakaike, H. Inoue, T. Suganuma, Moriyoshi Ohara","doi":"10.1109/IISWC.2014.6983055","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983055","url":null,"abstract":"Method inlining is a standard optimization in optimizing compilers. Java Just-In-Time (JIT) compilers use a size threshold for their inlining decisions. They favor inlining methods that are smaller than the threshold and do not inline the other larger methods. This has been a simple and effective strategy, but we should be able to make even better inlining based on a cost-benefit model, assuming accurate and lightweight call-graph profiles. We characterized the call-graph profiles of twelve SPECjvm2008 benchmarks with two metrics, method size and call frequency, to study the opportunities that make better tradeoff between performance benefits and compilation costs over the existing size-based inlining algorithm in the dynamic compilation environment. We found that inlining methods that are smaller than a size threshold is not always beneficial for performance. Most of the call-graph edges to the small methods had quite low call frequencies. Excluding the low-weighted call-graph edges from inlining can be an opportunity to reduce the compilation cost. On the other hand, a few call-graph edges to the methods that are larger than the threshold had high call frequencies. Inlining the highly-weighted call-graph edges can be an opportunity to improve performance. We developed a new cost-benefit-based inlining algorithm to capture those identified opportunities and to make inlining decisions more cost effective. Our evaluation showed that the new algorithm improves performance significantly while reduces the compilation cost, compared to the widely-used size-based inlining algorithm.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121763231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Deep-dive analysis of the data analytics workload in CloudSuite 深入分析CloudSuite中的数据分析工作负载
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983059
Ahmad Yasin, Y. Ben-Asher, A. Mendelson
{"title":"Deep-dive analysis of the data analytics workload in CloudSuite","authors":"Ahmad Yasin, Y. Ben-Asher, A. Mendelson","doi":"10.1109/IISWC.2014.6983059","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983059","url":null,"abstract":"Exponential growth of digital data has introduced massively-parallel systems, special orchestration layers, and new scale-out applications. While recent works suggest characteristics of scale-out workloads are different from those of traditional ones, their root causes are not understood. Such understanding is extremely important to improve efficiency; even a 1% performance gain for a core can have a large impact on the datacenter as a whole. This paper studies the characteristics of a Big Data Analytics (BDA) workload on a modern cloud server. It is intentionally focused on a single workload-platform in order to enable deep-dive analysis that aims to understand the root causes of the CPU bottlenecks which this paper identify. We choose the Data Analytics benchmark from CloudSuite [1] as a representative of a growing family of important applications. This paper describes a customization of a comprehensive threefold analysis method. The method consists of a System level, where sensitivity to system parameters is examined, as well as Application and Architectural levels; where bottlenecks are attributed back to the application and runtime codes, respectively. The paper also adopts a proof-by-optimization approach to prove bottlenecks' validity. Overall, 65% net speedup is measured with significant power reduction. The paper reveals that BDA workloads suffer from overheads related to managing the data rather than accessing the data. For example, Hash index lookup is found to be a key performance limiter. Inefficiencies leading to such unexpected behavior are demonstrated, including JVM selection and heavily unoptimized application code, both of which have a big impact. Suboptimal microarchitecture areas are demonstrated too, in addition to programming styles that limit exploitation of upcoming JVM and CPU parallelization features.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"95 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126263998","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 54
Workload synthesis: Generating benchmark workloads from statistical execution profile 工作负载合成:从统计执行概要文件生成基准工作负载
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983051
Keunsoo Kim, Changmin Lee, Jung Ho Jung, W. Ro
{"title":"Workload synthesis: Generating benchmark workloads from statistical execution profile","authors":"Keunsoo Kim, Changmin Lee, Jung Ho Jung, W. Ro","doi":"10.1109/IISWC.2014.6983051","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983051","url":null,"abstract":"We propose an approach for benchmark workload generation. The proposed workload synthesis generates synthetic workloads that model the behavior of real applications. Statistical execution profile of a workload is constructed from hardware performance counters available in recent processors, and the overhead of profiling is significantly lower than instrumentation or simulation which requires inspection of instruction stream. Workload synthesis can be applied even though the source codes or binaries of real applications are not available, because it utilizes only statistical profile. In addition, for non-deterministic workloads, using synthetic workloads provides more consistent results than executing real workloads, since a synthetic workload replays predetermined instructions reconstructed from the execution profile of a real workload. Furthermore, with a sampling technique, we can reduce the execution time of synthetic workloads while preserving its run-time characteristics. We have implemented and evaluated the proposed method on ARM-based mobile devices. The results show that synthetic workloads reproduce the profiled performance event counts of real workloads with high accuracy.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116425112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 8
Tradeoffs between power management and tail latency in warehouse-scale applications 仓库级应用程序中电源管理和尾部延迟之间的权衡
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983037
Svilen Kanev, K. Hazelwood, Gu-Yeon Wei, D. Brooks
{"title":"Tradeoffs between power management and tail latency in warehouse-scale applications","authors":"Svilen Kanev, K. Hazelwood, Gu-Yeon Wei, D. Brooks","doi":"10.1109/IISWC.2014.6983037","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983037","url":null,"abstract":"The growth in datacenter computing has increased the importance of energy-efficiency in servers. Techniques to reduce power have brought server designs close to achieving energy-proportional computing. However, they stress the inherent tradeoff between aggressive power management and quality of service (QoS) - the dominant metric of performance in datacenters. In this paper, we characterize this tradeoff for 15 benchmarks representing workloads from Google's datacenters. We show that 9 of these benchmarks often toggle their cores between short bursts of activity and sleep. In doing so, they stress sleep selection algorithms and can cause tail latency degradation or missed potential for power savings of up to 10% on important workloads like web search. However, improving sleep selection alone is not sufficient for large efficiency gains on current server hardware. To guide the direction needed for such large gains, we profile datacenter applications for susceptibility to dynamic voltage and frequency scaling (DVFS). We find the largest potential in DVFS which is cognizant of latency/power tradeoffs on a workload-per-workload basis.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129418061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 66
Automatic source code analysis of branch mispredictions 分支错误预测的自动源代码分析
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983045
Celal Öztürk, Ibrahim Burak Karsli, Resit Sendag
{"title":"Automatic source code analysis of branch mispredictions","authors":"Celal Öztürk, Ibrahim Burak Karsli, Resit Sendag","doi":"10.1109/IISWC.2014.6983045","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983045","url":null,"abstract":"After over two decades of extensive research on branch prediction, branch mispredictions are still an important performance/power bottleneck for today's aggressive processors. In our prior work, to further understand the causes for mispredictions, we presented a source-code based classification of branch mispredictions extending the prior work on predictor-specific classification. Since source-code analysis by hand is very time-consuming and not possible in some cases, in this paper, we develop methods in order to automatically identify the data structures for each branch instruction, which allows detailed source-code analysis at run-time. We show that our run-time method can successfully provide source-code analysis and classify more than 99% of the branch mispredictions.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"63 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114587639","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 2
Graph processing on GPUs: Where are the bottlenecks? gpu上的图形处理:瓶颈在哪里?
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983053
Qiumin Xu, Hyeran Jeon, M. Annavaram
{"title":"Graph processing on GPUs: Where are the bottlenecks?","authors":"Qiumin Xu, Hyeran Jeon, M. Annavaram","doi":"10.1109/IISWC.2014.6983053","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983053","url":null,"abstract":"Large graph processing is now a critical component of many data analytics. Graph processing is used from social networking Web sites that provide context-aware services from user connectivity data to medical informatics that diagnose a disease from a given set of symptoms. Graph processing has several inherently parallel computation steps interspersed with synchronization needs. Graphics processing units (GPUs) are being proposed as a power-efficient choice for exploiting the inherent parallelism. There have been several efforts to efficiently map graph applications to GPUs. However, there have not been many characterization studies that provide an in-depth understanding of the interaction between the GPGPU hardware components and graph applications that are mapped to execute on GPUs. In this study, we compiled 12 graph applications and collected the performance and utilization statistics of the core components of GPU while running the applications on both a cycle accurate simulator and a real GPU card. We present detailed application execution characteristics on GPUs. Then, we discuss and suggest several approaches to optimize GPU hardware for enhancing the graph application performance.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134510807","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 81
CortexSuite: A synthetic brain benchmark suite CortexSuite:一个合成大脑基准套件
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983043
Shelby Thomas, Chetan Gohkale, Enrico Tanuwidjaja, Tony Chong, David Lau, Saturnino Garcia, M. Taylor
{"title":"CortexSuite: A synthetic brain benchmark suite","authors":"Shelby Thomas, Chetan Gohkale, Enrico Tanuwidjaja, Tony Chong, David Lau, Saturnino Garcia, M. Taylor","doi":"10.1109/IISWC.2014.6983043","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983043","url":null,"abstract":"These days, many traditional end-user applications are said to “run fast enough” on existing machines, so the search continues for novel applications that can leverage the new capabilities of our evolving hardware. Foremost of these potential applications are those that are clustered around information processing capabilities that humans have today but are lacking in computers. The fact that brains can perform these computations serves as an existence proof that these applications are realizable. At the same time, we often discover that the human nervous system, with its 80 billion neurons, on some metrics, is more powerful and energy-efficient than today's machines. Both of these aspects make this class of applications a desirable target for an architectural benchmark suite, because there is evidence that these applications are both useful and computationally challenging. This paper details CortexSuite, a Synthetic Brain Benchmark Suite, which seeks to capture this workload. We classify and identify benchmarks within CortexSuite by analogy to the human neural processing function. We use the major lobes of the cerebral cortex as a model for the organization and classification of data processing algorithms. To be clear, our goal is not to emulate the brain at the level of the neuron, but rather to collect together synthetic, man-made algorithms that have similar function and have met with success in the real world. We consulted six world-class machine learning and computer vision researchers, who collectively hold 83,091 citations across their distinct subareas, asking them to identify newly emerging computationally-intensive algorithms or applications that are going to have a large impact over the next ten years. This is coupled with datasets that reflect the philosophy of practical use algorithms and are coded in “clean C” so as to make them accessible, analyzable, and usable for parallel and approximate compiler and architecture researchers alike.","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"310 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121188024","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 62
Workload characterization of server-side JavaScript 服务器端JavaScript的工作负载表征
2014 IEEE International Symposium on Workload Characterization (IISWC) Pub Date : 2014-12-15 DOI: 10.1109/IISWC.2014.6983035
T. Ogasawara
{"title":"Workload characterization of server-side JavaScript","authors":"T. Ogasawara","doi":"10.1109/IISWC.2014.6983035","DOIUrl":"https://doi.org/10.1109/IISWC.2014.6983035","url":null,"abstract":"Recently, scripting languages are becoming popular as languages to develop server-side applications. Modern JavaScript compilers significantly optimize JavaScript code, but their main targets are client-side Web applications. In this paper, we characterize the runtime behaviors of server workloads on an emerging JavaScript server-side framework, Node.js, comparing it to client-side JavaScript code. The runtime profile shows that a large amount (47.5% on average) of the total CPU time is spent in the V8 C++ library used for server workloads, while only 3.2% is for client-side programs. Such high CPU usage in the C++ code limits the performance improvements for server workloads, since recent changes in the performance of client-side workloads are due to optimized JavaScript code. Our analysis of complex calling contexts reveals that function calls to the V8 runtime from the server-side framework to handle JavaScript objects significantly contribute to the high CPU usage in the V8 library (up to 22.5% of the total time and 15.4% on average). We also show that the function calls to the V8 runtime from compiled JavaScript code are another source of the high CPU usage in the V8 library code (up to 24.5% of the total and 18.7% on average). Only a few JavaScript functions that implement the server-side framework API make these function calls to the V8 runtime and contribute to a large amount of the CPU time (up to 6.8% of the total).","PeriodicalId":448397,"journal":{"name":"2014 IEEE International Symposium on Workload Characterization (IISWC)","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2014-12-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123701210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 16
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书