Alexey Tumanov, T. Zhu, J. Park, M. Kozuch, Mor Harchol-Balter, G. Ganger
TetriSched is a scheduler that works in tandem with a calendaring reservation system to continuously re-evaluate the immediate-term scheduling plan for all pending jobs (including those with reservations and best-effort jobs) on each scheduling cycle. TetriSched leverages information supplied by the reservation system about jobs' deadlines and estimated runtimes to plan ahead in deciding whether to wait for a busy preferred resource type (e.g., machine with a GPU) or fall back to less preferred placement options. Plan-ahead affords significant flexibility in handling mis-estimates in job runtimes specified at reservation time. Integrated with the main reservation system in Hadoop YARN, TetriSched is experimentally shown to achieve significantly higher SLO attainment and cluster utilization than the best-configured YARN reservation and CapacityScheduler stack deployed on a real 256 node cluster.
{"title":"TetriSched: global rescheduling with adaptive plan-ahead in dynamic heterogeneous clusters","authors":"Alexey Tumanov, T. Zhu, J. Park, M. Kozuch, Mor Harchol-Balter, G. Ganger","doi":"10.1145/2901318.2901355","DOIUrl":"https://doi.org/10.1145/2901318.2901355","url":null,"abstract":"TetriSched is a scheduler that works in tandem with a calendaring reservation system to continuously re-evaluate the immediate-term scheduling plan for all pending jobs (including those with reservations and best-effort jobs) on each scheduling cycle. TetriSched leverages information supplied by the reservation system about jobs' deadlines and estimated runtimes to plan ahead in deciding whether to wait for a busy preferred resource type (e.g., machine with a GPU) or fall back to less preferred placement options. Plan-ahead affords significant flexibility in handling mis-estimates in job runtimes specified at reservation time. Integrated with the main reservation system in Hadoop YARN, TetriSched is experimentally shown to achieve significantly higher SLO attainment and cluster utilization than the best-configured YARN reservation and CapacityScheduler stack deployed on a real 256 node cluster.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"74954600","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}
Jean-Pierre Lozi, Baptiste Lepers, Justin R. Funston, Fabien Gaud, Vivien Quéma, Alexandra Fedorova
As a central part of resource management, the OS thread scheduler must maintain the following, simple, invariant: make sure that ready threads are scheduled on available cores. As simple as it may seem, we found that this invariant is often broken in Linux. Cores may stay idle for seconds while ready threads are waiting in runqueues. In our experiments, these performance bugs caused many-fold performance degradation for synchronization-heavy scientific applications, 13% higher latency for kernel make, and a 14-23% decrease in TPC-H throughput for a widely used commercial database. The main contribution of this work is the discovery and analysis of these bugs and providing the fixes. Conventional testing techniques and debugging tools are ineffective at confirming or understanding this kind of bugs, because their symptoms are often evasive. To drive our investigation, we built new tools that check for violation of the invariant online and visualize scheduling activity. They are simple, easily portable across kernel versions, and run with a negligible overhead. We believe that making these tools part of the kernel developers' tool belt can help keep this type of bug at bay.
{"title":"The Linux scheduler: a decade of wasted cores","authors":"Jean-Pierre Lozi, Baptiste Lepers, Justin R. Funston, Fabien Gaud, Vivien Quéma, Alexandra Fedorova","doi":"10.1145/2901318.2901326","DOIUrl":"https://doi.org/10.1145/2901318.2901326","url":null,"abstract":"As a central part of resource management, the OS thread scheduler must maintain the following, simple, invariant: make sure that ready threads are scheduled on available cores. As simple as it may seem, we found that this invariant is often broken in Linux. Cores may stay idle for seconds while ready threads are waiting in runqueues. In our experiments, these performance bugs caused many-fold performance degradation for synchronization-heavy scientific applications, 13% higher latency for kernel make, and a 14-23% decrease in TPC-H throughput for a widely used commercial database. The main contribution of this work is the discovery and analysis of these bugs and providing the fixes. Conventional testing techniques and debugging tools are ineffective at confirming or understanding this kind of bugs, because their symptoms are often evasive. To drive our investigation, we built new tools that check for violation of the invariant online and visualize scheduling activity. They are simple, easily portable across kernel versions, and run with a negligible overhead. We believe that making these tools part of the kernel developers' tool belt can help keep this type of bug at bay.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2016-04-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"80598621","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}
Peng Wang, Guangyu Sun, Song Jiang, Jian Ouyang, Shiding Lin, Chen Zhang, J. Cong
Various key-value (KV) stores are widely employed for data management to support Internet services as they offer higher efficiency, scalability, and availability than relational database systems. The log-structured merge tree (LSM-tree) based KV stores have attracted growing attention because they can eliminate random writes and maintain acceptable read performance. Recently, as the price per unit capacity of NAND flash decreases, solid state disks (SSDs) have been extensively adopted in enterprise-scale data centers to provide high I/O bandwidth and low access latency. However, it is inefficient to naively combine LSM-tree-based KV stores with SSDs, as the high parallelism enabled within the SSD cannot be fully exploited. Current LSM-tree-based KV stores are designed without assuming SSD's multi-channel architecture. To address this inadequacy, we propose LOCS, a system equipped with a customized SSD design, which exposes its internal flash channels to applications, to work with the LSM-tree-based KV store, specifically LevelDB in this work. We extend LevelDB to explicitly leverage the multiple channels of an SSD to exploit its abundant parallelism. In addition, we optimize scheduling and dispatching polices for concurrent I/O requests to further improve the efficiency of data access. Compared with the scenario where a stock LevelDB runs on a conventional SSD, the throughput of storage system can be improved by more than 4X after applying all proposed optimization techniques.
{"title":"An efficient design and implementation of LSM-tree based key-value store on open-channel SSD","authors":"Peng Wang, Guangyu Sun, Song Jiang, Jian Ouyang, Shiding Lin, Chen Zhang, J. Cong","doi":"10.1145/2592798.2592804","DOIUrl":"https://doi.org/10.1145/2592798.2592804","url":null,"abstract":"Various key-value (KV) stores are widely employed for data management to support Internet services as they offer higher efficiency, scalability, and availability than relational database systems. The log-structured merge tree (LSM-tree) based KV stores have attracted growing attention because they can eliminate random writes and maintain acceptable read performance. Recently, as the price per unit capacity of NAND flash decreases, solid state disks (SSDs) have been extensively adopted in enterprise-scale data centers to provide high I/O bandwidth and low access latency. However, it is inefficient to naively combine LSM-tree-based KV stores with SSDs, as the high parallelism enabled within the SSD cannot be fully exploited. Current LSM-tree-based KV stores are designed without assuming SSD's multi-channel architecture.\u0000 To address this inadequacy, we propose LOCS, a system equipped with a customized SSD design, which exposes its internal flash channels to applications, to work with the LSM-tree-based KV store, specifically LevelDB in this work. We extend LevelDB to explicitly leverage the multiple channels of an SSD to exploit its abundant parallelism. In addition, we optimize scheduling and dispatching polices for concurrent I/O requests to further improve the efficiency of data access. Compared with the scenario where a stock LevelDB runs on a conventional SSD, the throughput of storage system can be improved by more than 4X after applying all proposed optimization techniques.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75700387","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}
This paper proposes a new approach to determining the order of interdependent operations in a distributed system. The key idea behind our approach is to factor the task of tracking happens-before relationships out of components that comprise the system, and to centralize them in a separate event ordering service. This not only simplifies implementation of individual components by freeing them from having to propagate dependence information, but also enables dependence relationships to be maintained across multiple independent systems. A novel API enables the system to detect and take advantage of concurrency whenever possible by maintaining fine-grained information and binding events to a time order as late as possible. We demonstrate the benefits of this approach through several example applications, including a transactional key-value store, and an online graph store. Experiments show that our event ordering service scales well and has low overhead in practice.
{"title":"Kronos: the design and implementation of an event ordering service","authors":"Robert Escriva, Ayush Dubey, B. Wong, E. G. Sirer","doi":"10.1145/2592798.2592822","DOIUrl":"https://doi.org/10.1145/2592798.2592822","url":null,"abstract":"This paper proposes a new approach to determining the order of interdependent operations in a distributed system. The key idea behind our approach is to factor the task of tracking happens-before relationships out of components that comprise the system, and to centralize them in a separate event ordering service. This not only simplifies implementation of individual components by freeing them from having to propagate dependence information, but also enables dependence relationships to be maintained across multiple independent systems. A novel API enables the system to detect and take advantage of concurrency whenever possible by maintaining fine-grained information and binding events to a time order as late as possible. We demonstrate the benefits of this approach through several example applications, including a transactional key-value store, and an online graph store. Experiments show that our event ordering service scales well and has low overhead in practice.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"83935526","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}
Sriram Subramanian, S. Sundararaman, Nisha Talagala, A. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau
Snapshots are a common and heavily relied upon feature in storage systems. The high performance of flash-based storage systems brings new, more stringent, requirements for this classic capability. We present ioSnap, a flash optimized snapshot system. Through careful design exploiting common snapshot usage patterns and flash oriented optimizations, including leveraging native characteristics of Flash Translation Layers, ioSnap delivers low-overhead snapshots with minimal disruption to foreground traffic. Through our evaluation, we show that ioSnap incurs negligible performance overhead during normal operation, and that common-case operations such as snapshot creation and deletion incur little cost. We also demonstrate techniques to mitigate the performance impact on foreground I/O during intensive snapshot operations such as activation. Overall, ioSnap represents a case study of how to integrate snapshots into a modern, well-engineered flash-based storage system.
{"title":"Snapshots in a flash with ioSnap","authors":"Sriram Subramanian, S. Sundararaman, Nisha Talagala, A. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau","doi":"10.1145/2592798.2592825","DOIUrl":"https://doi.org/10.1145/2592798.2592825","url":null,"abstract":"Snapshots are a common and heavily relied upon feature in storage systems. The high performance of flash-based storage systems brings new, more stringent, requirements for this classic capability. We present ioSnap, a flash optimized snapshot system. Through careful design exploiting common snapshot usage patterns and flash oriented optimizations, including leveraging native characteristics of Flash Translation Layers, ioSnap delivers low-overhead snapshots with minimal disruption to foreground traffic. Through our evaluation, we show that ioSnap incurs negligible performance overhead during normal operation, and that common-case operations such as snapshot creation and deletion incur little cost. We also demonstrate techniques to mitigate the performance impact on foreground I/O during intensive snapshot operations such as activation. Overall, ioSnap represents a case study of how to integrate snapshots into a modern, well-engineered flash-based storage system.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76773458","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}
Xiaozhou Li, D. Andersen, M. Kaminsky, M. Freedman
Fast concurrent hash tables are an increasingly important building block as we scale systems to greater numbers of cores and threads. This paper presents the design, implementation, and evaluation of a high-throughput and memory-efficient concurrent hash table that supports multiple readers and writers. The design arises from careful attention to systems-level optimizations such as minimizing critical section length and reducing interprocessor coherence traffic through algorithm re-engineering. As part of the architectural basis for this engineering, we include a discussion of our experience and results adopting Intel's recent hardware transactional memory (HTM) support to this critical building block. We find that naively allowing concurrent access using a coarse-grained lock on existing data structures reduces overall performance with more threads. While HTM mitigates this slowdown somewhat, it does not eliminate it. Algorithmic optimizations that benefit both HTM and designs for fine-grained locking are needed to achieve high performance. Our performance results demonstrate that our new hash table design---based around optimistic cuckoo hashing---outperforms other optimized concurrent hash tables by up to 2.5x for write-heavy workloads, even while using substantially less memory for small key-value items. On a 16-core machine, our hash table executes almost 40 million insert and more than 70 million lookup operations per second.
{"title":"Algorithmic improvements for fast concurrent Cuckoo hashing","authors":"Xiaozhou Li, D. Andersen, M. Kaminsky, M. Freedman","doi":"10.1145/2592798.2592820","DOIUrl":"https://doi.org/10.1145/2592798.2592820","url":null,"abstract":"Fast concurrent hash tables are an increasingly important building block as we scale systems to greater numbers of cores and threads. This paper presents the design, implementation, and evaluation of a high-throughput and memory-efficient concurrent hash table that supports multiple readers and writers. The design arises from careful attention to systems-level optimizations such as minimizing critical section length and reducing interprocessor coherence traffic through algorithm re-engineering. As part of the architectural basis for this engineering, we include a discussion of our experience and results adopting Intel's recent hardware transactional memory (HTM) support to this critical building block. We find that naively allowing concurrent access using a coarse-grained lock on existing data structures reduces overall performance with more threads. While HTM mitigates this slowdown somewhat, it does not eliminate it. Algorithmic optimizations that benefit both HTM and designs for fine-grained locking are needed to achieve high performance.\u0000 Our performance results demonstrate that our new hash table design---based around optimistic cuckoo hashing---outperforms other optimized concurrent hash tables by up to 2.5x for write-heavy workloads, even while using substantially less memory for small key-value items. On a 16-core machine, our hash table executes almost 40 million insert and more than 70 million lookup operations per second.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"79772483","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}
Zhenyu Guo, C. Hong, Mao Yang, Dong Zhou, Lidong Zhou, Li Zhuang
Standard state-machine replication involves consensus on a sequence of totally ordered requests through, for example, the Paxos protocol. Such a sequential execution model is becoming outdated on prevalent multi-core servers. Highly concurrent executions on multi-core architectures introduce non-determinism related to thread scheduling and lock contentions, and fundamentally break the assumption in state-machine replication. This tension between concurrency and consistency is not inherent because the total-ordering of requests is merely a simplifying convenience that is unnecessary for consistency. Concurrent executions of the application can be decoupled with a sequence of consensus decisions through consensus on partial-order traces, rather than on totally ordered requests, that capture the non-deterministic decisions in one replica execution and to be replayed with the same decisions on others. The result is a new multi-core friendly replicated state-machine framework that achieves strong consistency while preserving parallelism in multi-thread applications. On 12-core machines with hyper-threading, evaluations on typical applications show that we can scale with the number of cores, achieving up to 16 times the throughput of standard replicated state machines.
{"title":"Rex: replication at the speed of multi-core","authors":"Zhenyu Guo, C. Hong, Mao Yang, Dong Zhou, Lidong Zhou, Li Zhuang","doi":"10.1145/2592798.2592800","DOIUrl":"https://doi.org/10.1145/2592798.2592800","url":null,"abstract":"Standard state-machine replication involves consensus on a sequence of totally ordered requests through, for example, the Paxos protocol. Such a sequential execution model is becoming outdated on prevalent multi-core servers. Highly concurrent executions on multi-core architectures introduce non-determinism related to thread scheduling and lock contentions, and fundamentally break the assumption in state-machine replication. This tension between concurrency and consistency is not inherent because the total-ordering of requests is merely a simplifying convenience that is unnecessary for consistency. Concurrent executions of the application can be decoupled with a sequence of consensus decisions through consensus on partial-order traces, rather than on totally ordered requests, that capture the non-deterministic decisions in one replica execution and to be replayed with the same decisions on others. The result is a new multi-core friendly replicated state-machine framework that achieves strong consistency while preserving parallelism in multi-thread applications. On 12-core machines with hyper-threading, evaluations on typical applications show that we can scale with the number of cores, achieving up to 16 times the throughput of standard replicated state machines.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76424484","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}
Haris Volos, Sanketh Nalli, S. Panneerselvam, V. Varadarajan, Prashant Saxena, M. Swift
Storage-class memory technologies such as phase-change memory and memristors present a radically different interface to storage than existing block devices. As a result, they provide a unique opportunity to re-examine storage architectures. We find that the existing kernel-based stack of components, well suited for disks, unnecessarily limits the design and implementation of file systems for this new technology. We present Aerie, a flexible file-system architecture that exposes storage-class memory to user-mode programs so they can access files without kernel interaction. Aerie can implement a generic POSIX-like file system with performance similar to or better than a kernel implementation. The main benefit of Aerie, though, comes from enabling applications to optimize the file system interface. We demonstrate a specialized file system that reduces a hierarchical file system abstraction to a key/value store with fewer consistency guarantees but 20-109% higher performance than a kernel file system.
{"title":"Aerie: flexible file-system interfaces to storage-class memory","authors":"Haris Volos, Sanketh Nalli, S. Panneerselvam, V. Varadarajan, Prashant Saxena, M. Swift","doi":"10.1145/2592798.2592810","DOIUrl":"https://doi.org/10.1145/2592798.2592810","url":null,"abstract":"Storage-class memory technologies such as phase-change memory and memristors present a radically different interface to storage than existing block devices. As a result, they provide a unique opportunity to re-examine storage architectures. We find that the existing kernel-based stack of components, well suited for disks, unnecessarily limits the design and implementation of file systems for this new technology.\u0000 We present Aerie, a flexible file-system architecture that exposes storage-class memory to user-mode programs so they can access files without kernel interaction. Aerie can implement a generic POSIX-like file system with performance similar to or better than a kernel implementation. The main benefit of Aerie, though, comes from enabling applications to optimize the file system interface. We demonstrate a specialized file system that reduces a hierarchical file system abstraction to a key/value store with fewer consistency guarantees but 20-109% higher performance than a kernel file system.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"88508219","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}
Jian Tan, Alicia Chin, Z. Z. Hu, Yonggang Hu, S. Meng, Xiaoqiao Meng, Li Zhang
In order to improve the performance of MapReduce, we design DynMR. It addresses the following problems that persist in the existing implementations: 1) difficulty in selecting optimal performance parameters for a single job in a fixed, dedicated environment, and lack of capability to configure parameters that can perform optimally in a dynamic, multi-job cluster; 2) long job execution resulting from a task long-tail effect, often caused by ReduceTask data skew or heterogeneous computing nodes; 3) inefficient use of hardware resources, since ReduceTasks bundle several functional phases together and may idle during certain phases. DynMR adaptively interleaves the execution of several partially-completed ReduceTasks and backfills MapTasks so that they run in the same JVM, one at a time. It consists of three components. 1) A running ReduceTask uses a detection algorithm to identify resource underutilization during the shuffle phase. It then gives up the allocated hardware resources efficiently to the next task. 2) A number of ReduceTasks are gradually assembled in a progressive queue, according to a flow control algorithm in runtime. These tasks execute in an interleaved rotation. Additional ReduceTasks can be inserted adaptively to the progressive queue if the full fetching capacity is not reached. MapTasks can be back-filled therein if it is still underused. 3) Merge threads of each ReduceTask are extracted out as standalone services within the associated JVM. This design allows the data segments of multiple partially-complete ReduceTasks to reside in the same JVM heap, controlled by a segment manager and served by the common merge threads. Experiments show 10% ~ 40% improvements, depending on the workload.
{"title":"DynMR: dynamic MapReduce with ReduceTask interleaving and MapTask backfilling","authors":"Jian Tan, Alicia Chin, Z. Z. Hu, Yonggang Hu, S. Meng, Xiaoqiao Meng, Li Zhang","doi":"10.1145/2592798.2592805","DOIUrl":"https://doi.org/10.1145/2592798.2592805","url":null,"abstract":"In order to improve the performance of MapReduce, we design DynMR. It addresses the following problems that persist in the existing implementations: 1) difficulty in selecting optimal performance parameters for a single job in a fixed, dedicated environment, and lack of capability to configure parameters that can perform optimally in a dynamic, multi-job cluster; 2) long job execution resulting from a task long-tail effect, often caused by ReduceTask data skew or heterogeneous computing nodes; 3) inefficient use of hardware resources, since ReduceTasks bundle several functional phases together and may idle during certain phases.\u0000 DynMR adaptively interleaves the execution of several partially-completed ReduceTasks and backfills MapTasks so that they run in the same JVM, one at a time. It consists of three components. 1) A running ReduceTask uses a detection algorithm to identify resource underutilization during the shuffle phase. It then gives up the allocated hardware resources efficiently to the next task. 2) A number of ReduceTasks are gradually assembled in a progressive queue, according to a flow control algorithm in runtime. These tasks execute in an interleaved rotation. Additional ReduceTasks can be inserted adaptively to the progressive queue if the full fetching capacity is not reached. MapTasks can be back-filled therein if it is still underused. 3) Merge threads of each ReduceTask are extracted out as standalone services within the associated JVM. This design allows the data segments of multiple partially-complete ReduceTasks to reside in the same JVM heap, controlled by a segment manager and served by the common merge threads. Experiments show 10% ~ 40% improvements, depending on the workload.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"86830590","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}
K. Zarifis, Rui Miao, Matt Calder, Ethan Katz-Bassett, Minlan Yu, J. Padhye
Data centers must support a range of workloads with differing demands. Although existing approaches handle routine traffic smoothly, intense hotspots--even if ephemeral--cause excessive packet loss and severely degrade performance. This loss occurs even though congestion is typically highly localized, with spare buffer capacity at nearby switches. In this paper, we argue that switches should share buffer capacity to effectively handle this spot congestion without the monetary hit of deploying large buffers at individual switches. Specifically, we present detour-induced buffer sharing (DIBS), a mechanism that achieves a near lossless network without requiring additional buffers at individual switches. Using DIBS, a congested switch detours packets randomly to neighboring switches to avoid dropping the packets. We implement DIBS in hardware, on software routers in a testbed, and in simulation, and we demonstrate that it reduces the 99th percentile of delay-sensitive query completion time by up to 85%, with very little impact on other traffic.
{"title":"DIBS: just-in-time congestion mitigation for data centers","authors":"K. Zarifis, Rui Miao, Matt Calder, Ethan Katz-Bassett, Minlan Yu, J. Padhye","doi":"10.1145/2592798.2592806","DOIUrl":"https://doi.org/10.1145/2592798.2592806","url":null,"abstract":"Data centers must support a range of workloads with differing demands. Although existing approaches handle routine traffic smoothly, intense hotspots--even if ephemeral--cause excessive packet loss and severely degrade performance. This loss occurs even though congestion is typically highly localized, with spare buffer capacity at nearby switches. In this paper, we argue that switches should share buffer capacity to effectively handle this spot congestion without the monetary hit of deploying large buffers at individual switches. Specifically, we present detour-induced buffer sharing (DIBS), a mechanism that achieves a near lossless network without requiring additional buffers at individual switches. Using DIBS, a congested switch detours packets randomly to neighboring switches to avoid dropping the packets. We implement DIBS in hardware, on software routers in a testbed, and in simulation, and we demonstrate that it reduces the 99th percentile of delay-sensitive query completion time by up to 85%, with very little impact on other traffic.","PeriodicalId":20737,"journal":{"name":"Proceedings of the Eleventh European Conference on Computer Systems","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2014-04-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"87155641","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：