URL: https://doi.org/10.1145/3230543.3230572
Bibtex Entry:
@inproceedings{2018-Kim-sigcomm, author = “Kim, Daehyeok and Memaripour, Amirsaman and Badam, Anirudh and Zhu, Yibo and Liu, Hongqiang Harry and Padhye, Jitu and Raindel, Shachar and Swanson, Steven and Sekar, Vyas and Seshan, Srinivasan”, title = “Hyperloop: Group-Based NIC-Offloading to Accelerate Replicated Transactions in Multi-Tenant Storage Systems”, year = “2018”, month = “August”, isbn = “9781450355674”, publisher = “Association for Computing Machinery”, address = “New York, NY, USA”, url = “https://doi.org/10.1145/3230543.3230572”, doi = “10.1145/3230543.3230572”, abstract = “Storage systems in data centers are an important component of large-scale online services. They typically perform replicated transactional operations for high data availability and integrity. Today, however, such operations suffer from high tail latency even with recent kernel bypass and storage optimizations, and thus affect the predictability of end-to-end performance of these services. We observe that the root cause of the problem is the involvement of the CPU, a precious commodity in multi-tenant settings, in the critical path of replicated transactions. In this paper, we present HyperLoop, a new framework that removes CPU from the critical path of replicated transactions in storage systems by offloading them to commodity RDMA NICs, with non-volatile memory as the storage medium. To achieve this, we develop new and general NIC offloading primitives that can perform memory operations on all nodes in a replication group while guaranteeing ACID properties without CPU involvement. We demonstrate that popular storage applications can be easily optimized using our primitives. Our evaluation results with microbenchmarks and application benchmarks show that HyperLoop can reduce 99th percentile latency ≈ 800X with close to 0\% CPU consumption on replicas.”, booktitle = “ACM SIGCOMM”, pages = “297–312”, numpages = “16”, keywords = “replicated transactions, NIC-offloading, distributed storage systems, RDMA”, location = “Budapest, Hungary”, category = “DC Network”, series = “SIGCOMM ‘18” }