Y. Won, Joontaek Oh, Jaemin Jung, Gyeongyeol Choi, Seongbae Son, J. Hwang, Sangyeun Cho
This work is dedicated to eliminating the overhead required for guaranteeing the storage order in the modern IO stack. The existing block device adopts a prohibitively expensive approach in ensuring the storage order among write requests: interleaving the write requests with Transfer-and-Flush. For exploiting the cache barrier command for flash storage, we overhaul the IO scheduler, the dispatch module, and the filesystem so that these layers are orchestrated to preserve the ordering condition imposed by the application with which the associated data blocks are made durable. The key ingredients of Barrier-Enabled IO stack are Epoch-based IO scheduling, Order-Preserving Dispatch, and Dual-Mode Journaling. Barrier-enabled IO stack can control the storage order without Transfer-and-Flush overhead. We implement the barrier-enabled IO stack in server as well as in mobile platforms. SQLite performance increases by 270% and 75%, in server and in smartphone, respectively. In a server storage, BarrierFS brings as much as by 43 × and by 73× performance gain in MySQL and SQLite, respectively, against EXT4 via relaxing the durability of a transaction.
{"title":"Bringing Order to Chaos","authors":"Y. Won, Joontaek Oh, Jaemin Jung, Gyeongyeol Choi, Seongbae Son, J. Hwang, Sangyeun Cho","doi":"10.1145/3242091","DOIUrl":"https://doi.org/10.1145/3242091","url":null,"abstract":"This work is dedicated to eliminating the overhead required for guaranteeing the storage order in the modern IO stack. The existing block device adopts a prohibitively expensive approach in ensuring the storage order among write requests: interleaving the write requests with Transfer-and-Flush. For exploiting the cache barrier command for flash storage, we overhaul the IO scheduler, the dispatch module, and the filesystem so that these layers are orchestrated to preserve the ordering condition imposed by the application with which the associated data blocks are made durable. The key ingredients of Barrier-Enabled IO stack are Epoch-based IO scheduling, Order-Preserving Dispatch, and Dual-Mode Journaling. Barrier-enabled IO stack can control the storage order without Transfer-and-Flush overhead. We implement the barrier-enabled IO stack in server as well as in mobile platforms. SQLite performance increases by 270% and 75%, in server and in smartphone, respectively. In a server storage, BarrierFS brings as much as by 43 × and by 73× performance gain in MySQL and SQLite, respectively, against EXT4 via relaxing the durability of a transaction.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"355 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114083302","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}
Haryadi S. Gunawi, Riza O. Suminto, R. Sears, Casey Golliher, S. Sundararaman, Xing Lin, Tim Emami, Weiguang Sheng, N. Bidokhti, C. McCaffrey, Deepthi Srinivasan, Biswaranjan Panda, A. Baptist, G. Grider, P. Fields, K. Harms, R. Ross, Andree Jacobson, R. Ricci, Kirk Webb, P. Alvaro, H. Runesha, M. Hao, Huaicheng Li
Fail-slow hardware is an under-studied failure mode. We present a study of 114 reports of fail-slow hardware incidents, collected from large-scale cluster deployments in 14 institutions. We show that all hardware types such as disk, SSD, CPU, memory, and network components can exhibit performance faults. We made several important observations such as faults convert from one form to another, the cascading root causes and impacts can be long, and fail-slow faults can have varying symptoms. From this study, we make suggestions to vendors, operators, and systems designers.
{"title":"Fail-Slow at Scale","authors":"Haryadi S. Gunawi, Riza O. Suminto, R. Sears, Casey Golliher, S. Sundararaman, Xing Lin, Tim Emami, Weiguang Sheng, N. Bidokhti, C. McCaffrey, Deepthi Srinivasan, Biswaranjan Panda, A. Baptist, G. Grider, P. Fields, K. Harms, R. Ross, Andree Jacobson, R. Ricci, Kirk Webb, P. Alvaro, H. Runesha, M. Hao, Huaicheng Li","doi":"10.1145/3242086","DOIUrl":"https://doi.org/10.1145/3242086","url":null,"abstract":"Fail-slow hardware is an under-studied failure mode. We present a study of 114 reports of fail-slow hardware incidents, collected from large-scale cluster deployments in 14 institutions. We show that all hardware types such as disk, SSD, CPU, memory, and network components can exhibit performance faults. We made several important observations such as faults convert from one form to another, the cascading root causes and impacts can be long, and fail-slow faults can have varying symptoms. From this study, we make suggestions to vendors, operators, and systems designers.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"50 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116437695","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}
Phase Change Memory (PCM) has drawn great attention as a main memory due to its attractive characteristics such as non-volatility, byte-addressability, and in-place update. However, since the capacity of PCM is not fully mature yet, hybrid memory architecture that consists of DRAM and PCM has been suggested as a main memory. In addition, page replacement algorithm based on hybrid memory architecture is actively being studied, because existing page replacement algorithms cannot be used on hybrid memory architecture in that they do not consider the two weaknesses of PCM: high write latency and low endurance. In this article, to mitigate the above hardware limitations of PCM, we revisit the page cache layer for the hybrid memory architecture and propose a novel page replacement algorithm, called M-CLOCK, to improve the performance of hybrid memory architecture and the lifespan of PCM. In particular, M-CLOCK aims to reduce the number of PCM writes that negatively affect the performance of hybrid memory architecture. Experimental results clearly show that M-CLOCK outperforms the state-of-the-art page replacement algorithms in terms of the number of PCM writes and effective memory access time by up to 98% and 9.4 times, respectively.
{"title":"M-CLOCK","authors":"Minhoe Lee, Donghyun Kang, Y. Eom","doi":"10.1145/3216730","DOIUrl":"https://doi.org/10.1145/3216730","url":null,"abstract":"Phase Change Memory (PCM) has drawn great attention as a main memory due to its attractive characteristics such as non-volatility, byte-addressability, and in-place update. However, since the capacity of PCM is not fully mature yet, hybrid memory architecture that consists of DRAM and PCM has been suggested as a main memory. In addition, page replacement algorithm based on hybrid memory architecture is actively being studied, because existing page replacement algorithms cannot be used on hybrid memory architecture in that they do not consider the two weaknesses of PCM: high write latency and low endurance. In this article, to mitigate the above hardware limitations of PCM, we revisit the page cache layer for the hybrid memory architecture and propose a novel page replacement algorithm, called M-CLOCK, to improve the performance of hybrid memory architecture and the lifespan of PCM. In particular, M-CLOCK aims to reduce the number of PCM writes that negatively affect the performance of hybrid memory architecture. Experimental results clearly show that M-CLOCK outperforms the state-of-the-art page replacement algorithms in terms of the number of PCM writes and effective memory access time by up to 98% and 9.4 times, respectively.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125505459","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}
R. Alagappan, Aishwarya Ganesan, Eric Lee, Aws Albarghouthi, Vijay Chidambaram, A. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau
We introduce protocol-aware recovery (Par), a new approach that exploits protocol-specific knowledge to correctly recover from storage faults in distributed systems. We demonstrate the efficacy of Par through the design and implementation of corruption-tolerant replication (Ctrl), a Par mechanism specific to replicated state machine (RSM) systems. We experimentally show that the Ctrl versions of two systems, LogCabin and ZooKeeper, safely recover from storage faults and provide high availability, while the unmodified versions can lose data or become unavailable. We also show that the Ctrl versions achieve this reliability with little performance overheads.
{"title":"Protocol-Aware Recovery for Consensus-Based Distributed Storage","authors":"R. Alagappan, Aishwarya Ganesan, Eric Lee, Aws Albarghouthi, Vijay Chidambaram, A. Arpaci-Dusseau, Remzi H. Arpaci-Dusseau","doi":"10.1145/3241062","DOIUrl":"https://doi.org/10.1145/3241062","url":null,"abstract":"We introduce protocol-aware recovery (Par), a new approach that exploits protocol-specific knowledge to correctly recover from storage faults in distributed systems. We demonstrate the efficacy of Par through the design and implementation of <underline>c</underline>orruption-<underline>t</underline>olerant <underline>r</underline>ep<underline>l</underline>ication (Ctrl), a Par mechanism specific to replicated state machine (RSM) systems. We experimentally show that the Ctrl versions of two systems, LogCabin and ZooKeeper, safely recover from storage faults and provide high availability, while the unmodified versions can lose data or become unavailable. We also show that the Ctrl versions achieve this reliability with little performance overheads.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131641295","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}
We present Lerna, an end-to-end tool that automatically and transparently detects and extracts parallelism from data-dependent sequential loops. Lerna uses speculation combined with a set of techniques including code profiling, dependency analysis, instrumentation, and adaptive execution. Speculation is needed to avoid conservative actions and detect actual conflicts. Lerna targets applications that are hard-to-parallelize due to data dependency. Our experimental study involves the parallelization of 13 applications with data dependencies. Results on a 24-core machine show an average of 2.7× speedup for micro-benchmarks and 2.5× for the macro-benchmarks.
{"title":"Lerna","authors":"Mohamed M. Saad, R. Palmieri, B. Ravindran","doi":"10.1145/3310368","DOIUrl":"https://doi.org/10.1145/3310368","url":null,"abstract":"We present Lerna, an end-to-end tool that automatically and transparently detects and extracts parallelism from data-dependent sequential loops. Lerna uses speculation combined with a set of techniques including code profiling, dependency analysis, instrumentation, and adaptive execution. Speculation is needed to avoid conservative actions and detect actual conflicts. Lerna targets applications that are hard-to-parallelize due to data dependency. Our experimental study involves the parallelization of 13 applications with data dependencies. Results on a 24-core machine show an average of 2.7× speedup for micro-benchmarks and 2.5× for the macro-benchmarks.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132780023","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 article presents REGISTOR, a platform for regular expression grabbing inside storage. The main idea of Registor is accelerating regular expression (regex) search inside storage where large data set is stored, eliminating the I/O bottleneck problem. A special hardware engine for regex search is designed and augmented inside a flash SSD that processes data on-the-fly during data transmission from NAND flash to host. To make the speed of regex search match the internal bus speed of a modern SSD, a deep pipeline structure is designed in Registor hardware consisting of a file semantics extractor, matching candidates finder, regex matching units (REMUs), and results organizer. Furthermore, each stage of the pipeline makes the use of maximal parallelism possible. To make Registor readily usable by high-level applications, we have developed a set of APIs and libraries in Linux allowing Registor to process files in the SSD by recombining separate data blocks into files efficiently. A working prototype of Registor has been built in our newly designed NVMe-SSD. Extensive experiments and analyses have been carried out to show that Registor achieves high throughput, reduces the I/O bandwidth requirement by up to 97%, and reduces CPU utilization by as much as 82% for regex search in large datasets.
{"title":"REGISTOR","authors":"Shuyi Pei, Jing Yang, Qing Yang","doi":"10.1145/3310149","DOIUrl":"https://doi.org/10.1145/3310149","url":null,"abstract":"This article presents REGISTOR, a platform for regular expression grabbing inside storage. The main idea of Registor is accelerating regular expression (regex) search inside storage where large data set is stored, eliminating the I/O bottleneck problem. A special hardware engine for regex search is designed and augmented inside a flash SSD that processes data on-the-fly during data transmission from NAND flash to host. To make the speed of regex search match the internal bus speed of a modern SSD, a deep pipeline structure is designed in Registor hardware consisting of a file semantics extractor, matching candidates finder, regex matching units (REMUs), and results organizer. Furthermore, each stage of the pipeline makes the use of maximal parallelism possible. To make Registor readily usable by high-level applications, we have developed a set of APIs and libraries in Linux allowing Registor to process files in the SSD by recombining separate data blocks into files efficiently. A working prototype of Registor has been built in our newly designed NVMe-SSD. Extensive experiments and analyses have been carried out to show that Registor achieves high throughput, reduces the I/O bandwidth requirement by up to 97%, and reduces CPU utilization by as much as 82% for regex search in large datasets.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116670237","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}
Zhen Sun, G. Kuenning, Sonam Mandal, Philip Shilane, Vasily Tarasov, Nong Xiao, E. Zadok
Deduplication has become essential in disk-based backup systems, but there have been few long-term studies of backup workloads. Most past studies either were of a small static snapshot or covered only a short period that was not representative of how a backup system evolves over time. For this article, we first collected 21 months of data from a shared user file system; 33 users and over 4,000 snapshots are covered. We then analyzed the dataset, examining a variety of essential characteristics across two dimensions: single-node deduplication and cluster deduplication. For single-node deduplication analysis, our primary focus was individual-user data. Despite apparently similar roles and behavior among all of our users, we found significant differences in their deduplication ratios. Moreover, the data that some users share with others had a much higher deduplication ratio than average. For cluster deduplication analysis, we implemented seven published data-routing algorithms and created a detailed comparison of their performance with respect to deduplication ratio, load distribution, and communication overhead. We found that per-file routing achieves a higher deduplication ratio than routing by super-chunk (multiple consecutive chunks), but it also leads to high data skew (imbalance of space usage across nodes). We also found that large chunking sizes are better for cluster deduplication, as they significantly reduce data-routing overhead, while their negative impact on deduplication ratios is small and acceptable. We draw interesting conclusions from both single-node and cluster deduplication analysis and make recommendations for future deduplication systems design.
{"title":"Cluster and Single-Node Analysis of Long-Term Deduplication Patterns","authors":"Zhen Sun, G. Kuenning, Sonam Mandal, Philip Shilane, Vasily Tarasov, Nong Xiao, E. Zadok","doi":"10.1145/3183890","DOIUrl":"https://doi.org/10.1145/3183890","url":null,"abstract":"Deduplication has become essential in disk-based backup systems, but there have been few long-term studies of backup workloads. Most past studies either were of a small static snapshot or covered only a short period that was not representative of how a backup system evolves over time. For this article, we first collected 21 months of data from a shared user file system; 33 users and over 4,000 snapshots are covered. We then analyzed the dataset, examining a variety of essential characteristics across two dimensions: single-node deduplication and cluster deduplication. For single-node deduplication analysis, our primary focus was individual-user data. Despite apparently similar roles and behavior among all of our users, we found significant differences in their deduplication ratios. Moreover, the data that some users share with others had a much higher deduplication ratio than average. For cluster deduplication analysis, we implemented seven published data-routing algorithms and created a detailed comparison of their performance with respect to deduplication ratio, load distribution, and communication overhead. We found that per-file routing achieves a higher deduplication ratio than routing by super-chunk (multiple consecutive chunks), but it also leads to high data skew (imbalance of space usage across nodes). We also found that large chunking sizes are better for cluster deduplication, as they significantly reduce data-routing overhead, while their negative impact on deduplication ratios is small and acceptable. We draw interesting conclusions from both single-node and cluster deduplication analysis and make recommendations for future deduplication systems design.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-05-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127644451","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}
Deng Zhou, Vania Fang, T. Xie, Wen Pan, R. Kesavan, Tony Lin, N. Patel
Since little has been reported in the literature concerning enterprise storage system file-level request scheduling, we do not have enough knowledge about how various scheduling factors affect performance. Moreover, we are in lack of a good understanding on how to enhance request scheduling to adapt to the changing characteristics of workloads and hardware resources. To answer these questions, we first build a request scheduler prototype based on WAFL®, a mainstream file system running on numerous enterprise storage systems worldwide. Next, we use the prototype to quantitatively measure the impact of various scheduling configurations on performance on a NetApp®'s enterprise-class storage system. Several observations have been made. For example, we discover that in order to improve performance, the priority of write requests and non-preempted restarted requests should be boosted in some workloads. Inspired by these observations, we further propose two scheduling enhancement heuristics called SORD (size-oriented request dispatching) and QATS (queue-depth aware time slicing). Finally, we evaluate them by conducting a wide range of experiments using workloads generated by SPC-1 and SFS2014 on both HDD-based and all-flash platforms. Experimental results show that the combination of the two can noticeably reduce average request latency under some workloads.
{"title":"Empirical Evaluation and Enhancement of Enterprise Storage System Request Scheduling","authors":"Deng Zhou, Vania Fang, T. Xie, Wen Pan, R. Kesavan, Tony Lin, N. Patel","doi":"10.1145/3193741","DOIUrl":"https://doi.org/10.1145/3193741","url":null,"abstract":"Since little has been reported in the literature concerning enterprise storage system file-level request scheduling, we do not have enough knowledge about how various scheduling factors affect performance. Moreover, we are in lack of a good understanding on how to enhance request scheduling to adapt to the changing characteristics of workloads and hardware resources. To answer these questions, we first build a request scheduler prototype based on WAFL®, a mainstream file system running on numerous enterprise storage systems worldwide. Next, we use the prototype to quantitatively measure the impact of various scheduling configurations on performance on a NetApp®'s enterprise-class storage system. Several observations have been made. For example, we discover that in order to improve performance, the priority of write requests and non-preempted restarted requests should be boosted in some workloads. Inspired by these observations, we further propose two scheduling enhancement heuristics called SORD (size-oriented request dispatching) and QATS (queue-depth aware time slicing). Finally, we evaluate them by conducting a wide range of experiments using workloads generated by SPC-1 and SFS2014 on both HDD-based and all-flash platforms. Experimental results show that the combination of the two can noticeably reduce average request latency under some workloads.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122163306","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}
Xiameng Hu, Xiaolin Wang, Lan Zhou, Yingwei Luo, Zhenlin Wang, C. Ding, Chencheng Ye
The reuse distance (least recently used (LRU) stack distance) is an essential metric for performance prediction and optimization of storage cache. Over the past four decades, there have been steady improvements in the algorithmic efficiency of reuse distance measurement. This progress is accelerating in recent years, both in theory and practical implementation. In this article, we present a kinetic model of LRU cache memory, based on the average eviction time (AET) of the cached data. The AET model enables fast measurement and use of low-cost sampling. It can produce the miss ratio curve in linear time with extremely low space costs. On storage trace benchmarks, AET reduces the time and space costs compared to former techniques. Furthermore, AET is a composable model that can characterize shared cache behavior through sampling and modeling individual programs or traces.
{"title":"Fast Miss Ratio Curve Modeling for Storage Cache","authors":"Xiameng Hu, Xiaolin Wang, Lan Zhou, Yingwei Luo, Zhenlin Wang, C. Ding, Chencheng Ye","doi":"10.1145/3185751","DOIUrl":"https://doi.org/10.1145/3185751","url":null,"abstract":"The reuse distance (least recently used (LRU) stack distance) is an essential metric for performance prediction and optimization of storage cache. Over the past four decades, there have been steady improvements in the algorithmic efficiency of reuse distance measurement. This progress is accelerating in recent years, both in theory and practical implementation. In this article, we present a kinetic model of LRU cache memory, based on the average eviction time (AET) of the cached data. The AET model enables fast measurement and use of low-cost sampling. It can produce the miss ratio curve in linear time with extremely low space costs. On storage trace benchmarks, AET reduces the time and space costs compared to former techniques. Furthermore, AET is a composable model that can characterize shared cache behavior through sampling and modeling individual programs or traces.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"20 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125772837","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}
The article presents an analysis of drive workloads from enterprise storage systems. The drive workloads are obtained from field return units from a cross-section of enterprise storage system vendors and thus provides a view of the workload characteristics over a wide spectrum of end-user applications. The workload parameters that have been characterized include transfer lengths, access patterns, throughput, and utilization. The study shows that reads are the dominant workload accounting for 80% of the accesses to the drive. Writes are dominated by short block random accesses while reads range from random to highly sequential. A trend analysis over the period 2010–2014 shows that the workload has remained fairly constant even as the capacities of the drives shipped has steadily increased. The study shows that the data stored on disk drives is relatively cold—on average less than 4% of the drive capacity is accessed in a given 2h interval.
{"title":"Workload Characterization for Enterprise Disk Drives","authors":"A. Kashyap","doi":"10.1145/3151847","DOIUrl":"https://doi.org/10.1145/3151847","url":null,"abstract":"The article presents an analysis of drive workloads from enterprise storage systems. The drive workloads are obtained from field return units from a cross-section of enterprise storage system vendors and thus provides a view of the workload characteristics over a wide spectrum of end-user applications. The workload parameters that have been characterized include transfer lengths, access patterns, throughput, and utilization. The study shows that reads are the dominant workload accounting for 80% of the accesses to the drive. Writes are dominated by short block random accesses while reads range from random to highly sequential. A trend analysis over the period 2010–2014 shows that the workload has remained fairly constant even as the capacities of the drives shipped has steadily increased. The study shows that the data stored on disk drives is relatively cold—on average less than 4% of the drive capacity is accessed in a given 2h interval.","PeriodicalId":273014,"journal":{"name":"ACM Transactions on Storage (TOS)","volume":"131 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2018-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114827710","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：