請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/81627完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 洪士灝(Shih-Hao Hung) | |
| dc.contributor.author | SHIAO-SZU CHEN | en |
| dc.contributor.author | 陳孝思 | zh_TW |
| dc.date.accessioned | 2022-11-24T09:24:54Z | - |
| dc.date.available | 2022-11-24T09:24:54Z | - |
| dc.date.copyright | 2021-09-11 | |
| dc.date.issued | 2021 | |
| dc.date.submitted | 2021-09-08 | |
| dc.identifier.citation | K. A., K. M., and A. D. G. FaSST: Fast, Scalable and Simple Distributed Transactions with TwoSided (RDMA) Datagram RPCs. Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI 16), pages 185–201, 2016. Microsoft Azure Function. https://docs.microsoft.com/azure/azure-functions/. I. E. Akkus, R. Chen, I. Rimac, M. Stein, K. Satzke, A. Beck, P. Aditya, and V. Hilt. SAND: Towards HighPerformance Serverless Computing. Proceedings of the 2018 USENIX Annual Technical Conference (ATC 18), page 923–935, 2018. L. Ao, L. Izhikevich, G. M. Voelker, and G. Porter. Sprocket: A Serverless Video Processing Framework. Proceedings of the ACM Symposium on Cloud Computing (SoCC 18), page 263–274, 2018. Apache Crail (incubating). https://crail.apache.org/. AWS Lambda. https://aws.amazon.com/lambda/. AWS S3. https://aws.amazon.com/s3/. D. BarcelonaPons, M. SánchezArtigas, G. París, P. Sutra, and P. GarcíaLópez. On the FaaS Track: Building Stateful Distributed Applications with Serverless Architectures. Proceedings of the 20th International Middleware Conference (Middleware 19), page 41–54, 2019. B. Burns, B. Grant, D. Oppenheimer, E. Brewer, and J. Wilkes. Borg Omega and Kubernetes. ACM queue: tomorrow’s computing today, 14:70–93, 2016. Ceph. https://ceph.io/. Cloudflare Workers. https://workers.cloudflare.com/. FUSE. https://www.kernel.org/doc/html/v5.7/filesystems/fuse.html. Kubernetes Container Storage Interface. https://kubernetes-csi.github.io/docs/. A. Dragojević, D. Narayanan, M. Castro, and O. Hodson. FaRM: Fast Remote Memory. Proceedings of the 11th USENIX Symposium on Networked Systems Design and Implementation (NSDI 14), page 401–414, 2014. Kubernetes DaemonSet. https://kubernetes.io/docs/concepts/workloads/controllers/daemonset/. J. E., P. Q., V. S., S. I., and R. B. Occupy the Cloud: Distributed Computing for the 99%. Proceedings of the 2017 Symposium on Cloud Computing (2017), page 445– 451, 2017. S. Fouladi, F. Romero, D. Iter, Q. Li, S. Chatterjee, C. Kozyrakis, M. Zaharia, and K. Winstein. From Laptop to Lambda: Outsourcing Everyday Jobs to Thousands of Transient Functional Containers. Proceedings of the 2019 USENIX Annual Technical Conference (ATC 19), page 475–488, 2019. S. Fouladi, R. S. Wahby, B. Shacklett, K. Balasubramaniam, W. Zeng, R. Bhalerao, A. Sivaraman, G. Porter, and K. Winstei. Encoding, Fast and Slow: LowLatency Video Processing Using Thousands of Tiny Threads. Proceedings of the 14th USENIX Symposium on Networked Systems Design and Implementation (NSDI 17), page 363–376, 2017. GCC, the GNU Compiler Collection. https://gcc.gnu.org/. Google Cloud Function. https://cloud.google.com/functions. IBM Cloud Functions. https://cloud.ibm.com/functions/. N. S. Islam, M. W. Rahman, J. J. R. Rajachandrasekar, H. Wang, H. Subramoni, C. Murthy, and D. K. Panda. High performance RDMAbased design of HDFS over InfiniBand. Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis (SC 12), 2012. [23] N. S. Islam, M. W. ur Rahman, X. Lu, and D. K. Panda. High Performance Design for HDFS with ByteAddressability of NVM and RDMA. Proceedings of the 2016 International Conference on Supercomputing (ICS 16), 2016. Y.J., I.J., and S.S. Orion: A Distributed File System for NonVolatile Main Memory and RDMA Capable Networks. Proceedings of the 17th USENIX Conference on File and Storage Technologies (FAST 19), pages 221–234, 2019. jnrfuse. https://github.com/SerCeMan/jnr-fuse. J. Jose, H. Subramoni, K. Kandalla, M. W. ur Rahman H. Wang, S. Narravula, and D. K. Panda. A Scalable Memcached Design for InfiniBand Clusters Using Hybrid Transports. 2012 12th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID 12), page 236–243, 2012. J. Jose, H. Subramoni, M. Luo, M. Zhang, J. Huang, M. W. Rahman, N. S. Islam, H. W. X. Ouyang, S. Sur, and D. K. Panda. Memcached Design on High Performance RDMA Capable Interconnects. 2011 International Conference on Parallel Processing (ICPP 11), pages 743–752, 2011. A. Kalia, M. Kaminsky, and D. G. Andersen. Using RDMA Efficiently for Key Value Services. Proceedings of the 2014 ACM SIGCOMM Conference on Appli cations, Technologies, Architectures, and Protocols for Computer Communication (SIGCOMM 14), page 295–306, 2014. Knative. https://knative.dev/. Enable the Ability to Customize the Mounted Volume for UserContainer. https://github.com/knative/serving/issues/2260. Knative Support for ReadMany Persistent Volume Claim. https://github.com/knative/serving/issues/4579. PersistentVolumeClaim Volumes Should Be Allowed in the Revision Spec. https://github.com/knative/serving/issues/2025. S. Kotni, A. Nayak, V. Ganapathy, and A. Basu. Faastlane: Accelerating Function asaService Workflows. Proceedings of the 2021 USENIX Annual Technical Conference (ATC 21), 2021. Kubernetes. https://kubernetes.io/. B. Li, Z. Ruan, W. Xiao, Y. Lu, Y. Xiong, A. Putnam, E. Chen, and L. Zhang. KV Direct: HighPerformance InMemory KeyValue Store with Programmable NIC. Proceedings of the 26th ACM Symposium on Operating Systems Principles (SOSP 17), page 137–152, 2017. libfuse. https://github.com/libfuse/libfuse. LLVM TableGen. https://llvm.org/docs/TableGen/. Y. Lu, J. Shu, Y. Chen, and T. Li. Octopus: An RDMAEnabled Distributed Persistent Memory File System. Proceedings of the 2017 USENIX Annual Technical Conference (ATC 17), pages 773–785, 2017. A.Mahgoub, K.Shankar, S.Mitra, A.Klimovic, S.Chaterji, and S.Bagchi.SONIC: Applicationaware Data Passing for Chained Serverless Applications. Proceedings of the 2021 USENIX Annual Technical Conference (ATC 21), page 1–15, 2021. Memcached. https://www.memcached.org/. Minio. https://min.io/. C. Mitchell, Y. Geng, and J. Li. Using OneSided RDMA Reads to Build a Fast, CPUEfficient KeyValue Store. Proceedings of the 2013 USENIX Annual Technical Conference (ATC 13), page 103–114, 2013. I. Müller, R. Marroquín, and G. Alonso. Lambada: Interactive Data Analytics on Cold Data Using Serverless Cloud Infrastructure. Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data (SIGMOD 20), 2020. J. Nelson, B. Holt, B. Myers, P. Briggs, L. Ceze, S. Kahan, and M. Oskin. Latency Tolerant Software Distributed Shared Memory. Proceedings of the 2015 USENIX Annual Technical Conference (ATC 15), page 291–305, 2015. Oracle Functions. https://www.oracle.com/cloud-native/functions/. M. Perron, R. C. Fernandez, D. DeWitt, and S. Madden. Starling: A Scalable Query Engine on Cloud Functions. Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data (SIGMOD 20), 2020. Q. Pu, S. Venkataraman, and I. Stoica. Serverless Data Analytics with Flint. 2018 IEEE 11th International Conference on Cloud Computing (IEEE CLOUD 18), 2018. Q. Pu, S. Venkataraman, and I. Stoica. Shuffling, Fast and Slow: Scalable Analytics on Serverless Infrastructure. Proceedings of the 16th USENIX Symposium on Networked Systems Design and Implementation (NSDI 19), 2019. Kubernetes Persistent Volume. https://kubernetes.io/docs/concepts/storage/persistent-volumes/. A.Rajgarhia and A.Gehani. Performance and Extension of User Space File Systems. Proceedings of the 2010 ACM Symposium on Applied Computing (SAC 10), 2010. A Remote Direct Memory Access Protocol Specification. https://datatracker.ietf.org/doc/html/rfc5040. Redis. https://redis.io/. RDMA over Converged Ethernet. https://en.wikipedia.org/wiki/RDMA_over_Converged_Ethernet. Kubernetes Storage Class. https://kubernetes.io/docs/concepts/storage/storage-classes/. S. Shillaker and P. Pietzutch. Faasm: Lightweight Isolation for Efficient Stateful Serverless Computing. Proceedings of the 2020 USENIX Annual Technical Conference (ATC 20), page 419–433, 2020. K. Shvachko, H. Kuang, S. Radia, and R. Chansler. The Hadoop Distributed File System. 2010 IEEE 26th Symposium on Mass Storage Systems and Technologies (MSST 10), 2010. Apache Spark in AWS Lambda. https://github.com/qubole/spark-on-lambda. V. Sreekanti, C. Wu, X. C. Lin, J. M. Faleiro, J. E. Gonzalez, J. M. Hellerstein, and A. Tumanov. Cloudburst: Stateful FunctionsasaService. Proceedings of the VLDB Endowment, 13:2438–2452, 2020. J. Stender, B. Kolbeck, M. H. gqvist, and F. H. Babudb. To FUSE or Not to FUSE: Performance of UserSpace File Systems. Proceedings of the 15th USENIX Conference on File and Storage Technologies (FAST 17), 2017. P. Stuedi, A. Trivedi, and B. Metzler. Wimpy Nodes with 10GbE: Leveraging One Sided Operations in SoftRDMA to Boost Memcached. Proceedings of the 2012 USENIX Annual Technical Conference (ATC 12), page 347–353, 2012. P. Stuedi, A. Trivedi, J. Pfefferle, A. Klimovic, A. Schuepbach, and B. Metzle. Unification of Temporary Storage in the NodeKernel Architecture. Proceedings of the 2019 USENIX Annual Technical Conference (ATC 19), page 767–782, 2019. H. Tang, J. Zhang, and F. Zhang. Accelerating Ceph with RDMA and NVMeoF. 2018 Annual Open Fabrics Alliance Workshop, 2018. V. Tarasov, A. Gupta, K. Sourav, S. Trehan, and E. Zadok. Terra Incognita: On the Practicality of UserSpace File Systems. Proceedings of the 7th USENIX Workshop on Hot Topics in Storage (HotStorage 15), 2015. RDMA Protocol Verbs Specification. https://datatracker.ietf.org/doc/html/draft-hilland-rddp-verbs-00 Linux Virtual File System. https://www.kernel.org/doc/html/latest/filesystems/vfs.html Y. Wang, L. Zhang, J. Tan, M. Li, Y. Gao, X. Guerin, X. Meng, and S. Meng. Hydradb: A Resilient RDMAdriven KeyValue Middleware for InMemory Cluster Computing. Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC 15), pages 33–48, 2015. X. Wei, J. Shi, Y. Chen, R. Chen, and H. Chen. Fast and General Distributed Trans actions Using RDMA and HTM. Proceedings of the 11th European Conference on Computer Systems (EuroSys 16), page 1–17, 2016. X. Wei, J. Shi, Y. Chen, R. Chen, and H. Chen. Fast InMemory Transaction Pro cessing Using RDMA and HTM. ACM Transactions on Computer Systems, page 1–37, 2017. S. Weil, S. Brandt, E. Miller, D. Long, and C. Maltzahnn. Ceph: A Scalable, High Performance Distributed File System. Proceedings of the 7th USENIX Symposium on Operating Systems Design and Implementation (OSDI 06), page 307–320, 2006. Kubernetes, Serverless, and You (Cloud Next ’18). https://www.youtube.com/watch?v=LtELzpw1l1M. W. X., D. Z., C. R., and C. H. Deconstructing RDMAenabled Distributed Transactions: Hybrid is Better! Proceedings of the 13th USENIX Symposium on Operating Systems Design and Implementation (OSDI 18), pages 233–251, 2018. T. Zhang, D. Xie, F. Li, and R. Stutsman. Narrowing the Gap Between Serverless and its State with Storage Functions. Proceedings of the ACM Symposium on Cloud Computing (SoCC 19), page 1–12, 2019. A. D. ́cand D. Narayananand E. Nightingale, M. Renzelmann, A. Shami, A. Badam, and M. Castro. No Compromises: Distributed Transactions with Consistency, Avail ability, and Performance. Proceedings of the 25th ACM Symposium on Operating Systems Principles (SOSP 15), page 54–70, 2015. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/81627 | - |
| dc.description.abstract | 無伺服器架構的出現讓開發者不需要維護大量電腦組成的龐大叢集，依然可以在有需求時有可以無限自動擴展的無伺服器涵式來使用，且只需以毫秒為單位來付出這些涵式在執行時所使用資源的費用，因此許多研究嘗試打造使用大量的無伺服器涵式來平行執行的應用程式以減少執行時間。然而無伺服器架構雖然可以減少應用程式的計算時間，卻因為涵式之間的資料交換必須透過遠端資料庫而增加資料輸出與輸入的時間。 為了減少資料交換所需時間，我們提出一種使用遠端直接記憶體存取 (RDMA) 來加速存取遠端資料庫的方法。我們選擇的無伺服器平台是建構在 Kubernetes 上的 Knative 以避免受限於特定雲端廠商。我們的方法是基於 Apache Crail。 Apache Crail 是一個使用遠端直接記憶體存取的開源分散式資料庫，應用程式想要存取 Apache Crail 之前，必須先跟 Apache Crail 建立 Crail 客戶端與遠端直接記憶體存取 (RDMA) Queue Pair，這個過程非常花時間。然而因為無伺服器涵式是無狀態的，無伺服器涵式間不能共享 Crail 客戶端與遠端直接記憶體存取 (RDMA) Queue Pair，導致每一個涵式都必須創建 Crail 客戶端與遠端直接記憶體存取 (RDMA) Queue Pair，增加許多額外的執行時間。 為了解決上述問題，我們打造基於 Apache Crail 的 Kubernetes 永久性資料儲存卷 (PersistentVolume) 來讓不同無伺服器涵式共用 Crail 客戶端與遠端直接記憶體存取 (RDMA) Queue Pair 以節省重複創建與銷毀 Crail 客戶端與遠端直接記憶體存取 (RDMA) Queue Pair 的時間。 | zh_TW |
| dc.description.provenance | Made available in DSpace on 2022-11-24T09:24:54Z (GMT). No. of bitstreams: 1 U0001-0608202109262700.pdf: 7266443 bytes, checksum: a91ab000737ad9765090637f247a59a0 (MD5) Previous issue date: 2021 | en |
| dc.description.tableofcontents | 摘要 vii Abstract ix 1 Introduction 1 2 Related Works 5 2.1 ServerlessComputing 5 2.1.1 Knative 6 2.2 DataExchangeAmongServerlessFunctions 6 2.3 RemoteDataStoreWithRDMA 7 3 Methodology 9 3.1 Overview 9 3.2 KnativeSupport 10 3.3 DesignandImplementationofCrailPV 11 3.3.1 CrailFUSE 12 3.3.2 CrailCSIdriver 14 3.3.3 DeploymentandOperationinKubernetes 17 3.4 KnativeApplicationExampleUsingCrailPV 17 4 Evaluation 19 4.1 ExperimentalSetup 19 4.2 Microbenchmarks 21 4.2.1 CrailPV(RDMA)vsCrailClient 21 4.2.2 CrailPV(NVMf)vsMinIO 22 4.2.3 CrailPV(RDMA)vsCrailPV(TCP) 23 4.2.4 CrailPVvsRedis 23 4.3 RealWorldApplications 24 4.4 QualitativeComparison 26 5 Conclusion and Future Works 29 Bibliography 31 | |
| dc.language.iso | en | |
| dc.subject | Apache Crail | zh_TW |
| dc.subject | 無伺服器架構 | zh_TW |
| dc.subject | Kubernetes | zh_TW |
| dc.subject | Knative | zh_TW |
| dc.subject | 遠端直接記憶體存取 | zh_TW |
| dc.subject | Apache Crail | en |
| dc.subject | RDMA | en |
| dc.subject | Knative | en |
| dc.subject | Kubernetes | en |
| dc.subject | Serverless | en |
| dc.title | 使用遠端直接記憶體存取加速 Kubernetes 中 Knative 函式間的資料交換 | zh_TW |
| dc.title | Accelerate Data Exchange via RDMA for Knative Functions in Kubernetes | en |
| dc.date.schoolyear | 109-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 楊佳玲(Hsin-Tsai Liu),施吉昇(Chih-Yang Tseng),張原豪,梁文耀 | |
| dc.subject.keyword | 無伺服器架構,Kubernetes,Knative,遠端直接記憶體存取,Apache Crail, | zh_TW |
| dc.subject.keyword | Apache Crail,Serverless,Kubernetes,Knative,RDMA, | en |
| dc.relation.page | 38 | |
| dc.identifier.doi | 10.6342/NTU202102137 | |
| dc.rights.note | 未授權 | |
| dc.date.accepted | 2021-09-08 | |
| dc.contributor.author-college | 電機資訊學院 | zh_TW |
| dc.contributor.author-dept | 資訊工程學研究所 | zh_TW |
| 顯示於系所單位： | 資訊工程學系 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| U0001-0608202109262700.pdf 未授權公開取用 | 7.1 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。