FaRM: Fast Remote Memory

By A. Dragojevic, Dushyanth Narayanan, Miguel Oom Temudo de Castro, O. Hodson. Symposium on Networked Systems Design and Implementation, 2014

958 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/ab2bdfc3ca2e74eea7989f38276d54cc3a3b17d1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Design Guidelines for High Performance RDMA Systems (2016, methodology): “For example, clients can access a key-value store either by READing directly from the server’s memory [24, 13, 30, 28], or via RPCs as in HERD [20].”
Cloud Programming Simplified: A Berkeley View on Serverless Computing (2019, background): “ resources allocated by a process when the process ﬁnishes (or crashes). Furthermore, such storage must provide access protection and performance isolation across applications. RAMCloud [54] and FaRM [55] show that it is possible to build in-memory storage systems that can provide microsecond level ”
Congestion Control for Large-Scale RDMA Deployments (2015, background): “Traditional TCP/IP stacks have far higher latency [10].”
The RAMCloud Storage System (2015, background): “MICA and FaRM were already mentioned in Section 12.1.”
IX: A Protected Dataplane Operating System for High Throughput and Low Latency (2014, methodology): “Interrupt co-alescing (used to reduce processing overheads), queuing latency due to device driver processing intervals, the use of intermediate buffering, and CPU scheduling delays frequently add up to several hundred µ s of latency to remote requests.”
Datacenter RPCs can be General and Fast (2018, methodology): “Co-design with RDMA is popular, with numerous examples from key-value stores [44, 36, 24, 61, 60], state machine replication [52], and transaction processing systems [25, 61, 20].”
Arrakis (2014, background): “While RDMA provides the performance benefits of user-level networking to parallel applications, it is challenging to apply the model to a broader class of clientserver applications [21].”
Faasm: Lightweight Isolation for Efficient Stateful Serverless Computing (2020, unknown): “FaRM [24] and RAM-”
Shenango: Achieving High CPU Efficiency for Latency-sensitive Datacenter Workloads (2019, background): “Examples include MICA [46], IX [18], Arrakis [59], mTCP [35], Sandstorm [50], FaRM [25], HERD [39], RAMCloud [57], SoftNIC [31], ZygOS [61], Shinjuku [38], and eRPC [40].”
Efficient Memory Disaggregation with Infiniswap (2017, methodology): “With the advent of RDMA, there has been a renewed interest in DSM research, especially via the key-value interface [35, 51, 60, 63, 66, 69].”
Fast in-memory transaction processing using RDMA and HTM (2015, methodology): “To bridge HTM (which essentially uses OCC) and 2PL, DrTM implements the exclusive and shared locks using one-sided RDMA operations, which are cache-coherent with local accesses and thus provide strong consistency with HTM.”
Using RDMA efficiently for key-value services (2014, methodology): “In Section 5, we evaluate our system on a cluster of 187 nodes and compare it against FaRM [8] and Pilaf [21].”
Homa: a receiver-driven low-latency transport protocol using network priorities (2018, background): “Even a single connection for each application-server pair is problematic for large-scale applications ([22] § 3.1, [11] § 3.1), so it is probably not realistic to use multiple connections.”
No compromises: distributed transactions with consistency, availability, and performance (2015, methodology): “A “distributed uninterruptible power supply (UPS)” exploits the wide availability of Lithium-ion batteries to lower the cost of a data center UPS over a traditional, centralized approach that uses lead-acid batteries.”
NetCache: Balancing Key-Value Stores with Fast In-Network Caching (2017, background): “Given the high-throughput and low-latency requirements of large-scale Internet services, key-value storage systems are shifting to in-memory designs [1, 15, 16, 22, 23, 26, 27, 29, 30, 34, 36, 38, 42, 43].”