Junhua Fang, Rong Zhang, T. Fu, Zhenjie Zhang, Aoying Zhou, Junhua Zhu
Key-based workload partitioning is a common strategy used in parallel stream processing engines, enabling effective key-value tuple distribution over worker threads in a logical operator. It is likely to generate poor balancing performance when workload variance occurs on the incoming data stream. This paper presents a new key-based workload partitioning framework, with practical algorithms to support dynamic workload assignment for stateful operators. The framework combines hash-based and explicit key-based routing strategies for workload distribution, which specifies the destination worker threads for a handful of keys and assigns the other keys with the hash function. When short-term distribution fluctuations occur to the incoming data stream, the system adaptively updates the routing table containing the chosen keys, in order to rebalance the workload with minimal migration overhead within the stateful operator. We formulate the rebalance operation as an optimization problem, with multiple objectives on minimizing state migration costs, controlling the size of the routing table and breaking workload imbalance among worker threads. Despite of the NP-hardness nature behind the optimization formulation, we carefully investigate and justify the heuristics behind key (re)routing and state migration, to facilitate fast response to workload variance with ignorable cost to the normal processing in the distributed system. Empirical studies on synthetic data and real-world stream applications validate the usefulness of our proposals.
{"title":"Parallel Stream Processing Against Workload Skewness and Variance","authors":"Junhua Fang, Rong Zhang, T. Fu, Zhenjie Zhang, Aoying Zhou, Junhua Zhu","doi":"10.1145/3078597.3078613","DOIUrl":"https://doi.org/10.1145/3078597.3078613","url":null,"abstract":"Key-based workload partitioning is a common strategy used in parallel stream processing engines, enabling effective key-value tuple distribution over worker threads in a logical operator. It is likely to generate poor balancing performance when workload variance occurs on the incoming data stream. This paper presents a new key-based workload partitioning framework, with practical algorithms to support dynamic workload assignment for stateful operators. The framework combines hash-based and explicit key-based routing strategies for workload distribution, which specifies the destination worker threads for a handful of keys and assigns the other keys with the hash function. When short-term distribution fluctuations occur to the incoming data stream, the system adaptively updates the routing table containing the chosen keys, in order to rebalance the workload with minimal migration overhead within the stateful operator. We formulate the rebalance operation as an optimization problem, with multiple objectives on minimizing state migration costs, controlling the size of the routing table and breaking workload imbalance among worker threads. Despite of the NP-hardness nature behind the optimization formulation, we carefully investigate and justify the heuristics behind key (re)routing and state migration, to facilitate fast response to workload variance with ignorable cost to the normal processing in the distributed system. Empirical studies on synthetic data and real-world stream applications validate the usefulness of our proposals.","PeriodicalId":436194,"journal":{"name":"Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125873751","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}
Many distributed systems require coordination between the components involved. With the steady growth of such systems, the probability of failures increases, which necessitates scalable fault-tolerant agreement protocols. The most common practical agreement protocol, for such scenarios, is leader-based atomic broadcast. In this work, we propose AllConcur, a distributed system that provides agreement through a leaderless concurrent atomic broadcast algorithm, thus, not suffering from the bottleneck of a central coordinator. In AllConcur, all components exchange messages concurrently through a logical overlay network that employs early termination to minimize the agreement latency. Our implementation of AllConcur supports standard sockets-based TCP as well as high-performance InfiniBand Verbs communications. AllConcur can handle up to 135 million requests per second and achieves 17x higher throughput than today's standard leader-based protocols, such as Libpaxos. Thus, AllConcur is highly competitive with regard to existing solutions and, due to its decentralized approach, enables hitherto unattainable system designs in a variety of fields.
{"title":"AllConcur: Leaderless Concurrent Atomic Broadcast","authors":"Marius Poke, T. Hoefler, C. W. Glass","doi":"10.1145/3078597.3078598","DOIUrl":"https://doi.org/10.1145/3078597.3078598","url":null,"abstract":"Many distributed systems require coordination between the components involved. With the steady growth of such systems, the probability of failures increases, which necessitates scalable fault-tolerant agreement protocols. The most common practical agreement protocol, for such scenarios, is leader-based atomic broadcast. In this work, we propose AllConcur, a distributed system that provides agreement through a leaderless concurrent atomic broadcast algorithm, thus, not suffering from the bottleneck of a central coordinator. In AllConcur, all components exchange messages concurrently through a logical overlay network that employs early termination to minimize the agreement latency. Our implementation of AllConcur supports standard sockets-based TCP as well as high-performance InfiniBand Verbs communications. AllConcur can handle up to 135 million requests per second and achieves 17x higher throughput than today's standard leader-based protocols, such as Libpaxos. Thus, AllConcur is highly competitive with regard to existing solutions and, due to its decentralized approach, enables hitherto unattainable system designs in a variety of fields.","PeriodicalId":436194,"journal":{"name":"Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing","volume":"115 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132731872","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}
Disa Mhembere, Da Zheng, C. Priebe, J. Vogelstein, R. Burns
k-means is one of the most influential and utilized machine learning algorithms. Its computation limits the performance and scalability of many statistical analysis and machine learning tasks. We rethink and optimize k-means in terms of modern NUMA architectures to develop a novel parallelization scheme that delays and minimizes synchronization barriers. The k-means NUMA Optimized Routine knor) library has (i) in-memory knori), (ii) distributed memory (knord), and (ii) semi-external memory (textsf{knors}) modules that radically improve the performance of k-means for varying memory and hardware budgets. knori boosts performance for single machine datasets by an order of magnitude or more. textsf{knors} improves the scalability of k-means on a memory budget using SSDs. knors scales to billions of points on a single machine, using a fraction of the resources that distributed in-memory systems require. knord retains knori's performance characteristics, while scaling in-memory through distributed computation in the cloud. knor modifies Elkan's triangle inequality pruning algorithm such that we utilize it on billion-point datasets without the significant memory overhead of the original algorithm. We demonstrate knor outperforms distributed commercial products like H2O, Turi (formerly Dato, GraphLab) and Spark's MLlib by more than an order of magnitude for datasets of 107 to 109 points.
{"title":"knor: A NUMA-Optimized In-Memory, Distributed and Semi-External-Memory k-means Library","authors":"Disa Mhembere, Da Zheng, C. Priebe, J. Vogelstein, R. Burns","doi":"10.1145/3078597.3078607","DOIUrl":"https://doi.org/10.1145/3078597.3078607","url":null,"abstract":"k-means is one of the most influential and utilized machine learning algorithms. Its computation limits the performance and scalability of many statistical analysis and machine learning tasks. We rethink and optimize k-means in terms of modern NUMA architectures to develop a novel parallelization scheme that delays and minimizes synchronization barriers. The k-means NUMA Optimized Routine knor) library has (i) in-memory knori), (ii) distributed memory (knord), and (ii) semi-external memory (textsf{knors}) modules that radically improve the performance of k-means for varying memory and hardware budgets. knori boosts performance for single machine datasets by an order of magnitude or more. textsf{knors} improves the scalability of k-means on a memory budget using SSDs. knors scales to billions of points on a single machine, using a fraction of the resources that distributed in-memory systems require. knord retains knori's performance characteristics, while scaling in-memory through distributed computation in the cloud. knor modifies Elkan's triangle inequality pruning algorithm such that we utilize it on billion-point datasets without the significant memory overhead of the original algorithm. We demonstrate knor outperforms distributed commercial products like H2O, Turi (formerly Dato, GraphLab) and Spark's MLlib by more than an order of magnitude for datasets of 107 to 109 points.","PeriodicalId":436194,"journal":{"name":"Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing","volume":"162 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-06-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115999271","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}
{"title":"Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing","authors":"","doi":"10.1145/3078597","DOIUrl":"https://doi.org/10.1145/3078597","url":null,"abstract":"","PeriodicalId":436194,"journal":{"name":"Proceedings of the 26th International Symposium on High-Performance Parallel and Distributed Computing","volume":"44 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123372307","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}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：