An official website of the United States government
It is becoming increasingly popular for distributed systems to exploit offload to reduce load on the CPU. Remote Direct Memory Access (RDMA) offload, in particular, has become popular. However, RDMA still requires CPU intervention for complex offloads that go beyond simple remote memory access. As such, the offload potential is limited and RDMA-based systems usually have to work around such limitations. We present RedN, a principled, practical approach to implementing complex RDMA offloads, without requiring any hardware modifications. Using self-modifying RDMA chains, we lift the existing RDMA verbs interface to a Turing complete set of programming abstractions. We explore what is possible in terms of offload complexity and performance with a commodity RDMA NIC. We show how to integrate these RDMA chains into applications, such as the Memcached key-value store, allowing us to offload complex tasks such as key lookups. RedN can reduce the latency of key-value get operations by up to 2.6× compared to state-of-the-art KV designs that use one-sided RDMA primitives (e.g., FaRM-KV), as well as traditional RPC-over-RDMA approaches. Moreover, compared to these baselines, RedN provides performance isolation and, in the presence of contention, can reduce latency by up to 35× while providing applications with failure resiliency to OS and process crashes.
more »
« less
RDMA vs. RPC for Implementing Distributed Data Structures
Distributed data structures are key to implementing scalable applications for scientific simulations and data analysis. In this paper we look at two implementation styles for distributed data structures: remote direct memory access (RDMA) and remote procedure call (RPC). We focus on operations that require individual accesses to remote portions of a distributed data structure, e.g., accessing a hash table bucket or distributed queue, rather than global operations in which all processors collectively exchange information. We look at the trade-offs between the two styles through microbenchmarks and a performance model that approximates the cost of each. The RDMA operations have direct hardware support in the network and therefore lower latency and overhead, while the RPC operations are more expressive but higher cost and can suffer from lack of attentiveness from the remote side. We also run experiments to compare the real-world performance of RDMA- and RPC-based data structure operations with the predicted performance to evaluate the accuracy of our model, and show that while the model does not always precisely predict running time, it allows us to choose the best implementation in the examples shown. We believe this analysis will assist developers in designing data structures that will perform well on current network architectures, as well as network architects in providing better support for this class of distributed data structures.
more »
« less
- Award ID(s):
- 1823037
- PAR ID:
- 10171729
- Date Published:
- Journal Name:
- Proceedings of the IEEE/ACM 9th Workshop on Irregular Applications: Architectures and Algorithms
- Page Range / eLocation ID:
- 17–22
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
null (Ed.)Networkswith Remote DirectMemoryAccess (RDMA) support are becoming increasingly common. RDMA, however, offers a limited programming interface to remote memory that consists of read, write and atomic operations. With RDMA alone, completing the most basic operations on remote data structures often requires multiple round-trips over the network. Data-intensive systems strongly desire higher-level communication abstractions that supportmore complex interaction patterns. A natural candidate to consider is MPI, the de facto standard for developing high-performance applications in the HPC community. This paper critically evaluates the communication primitives of MPI and shows that using MPI in the context of a data processing system comes with its own set of insurmountable challenges. Based on this analysis, we propose a new communication abstraction named RDMO, or Remote DirectMemory Operation, that dispatches a short sequence of reads, writes and atomic operations to remote memory and executes them in a single round-trip.more » « less
-
Data centers are increasingly equipped with RDMAs. These network interfaces mark the advent of a new distributed system model where a node can directly access the remote memory of another. They have enabled microsecond-scale replicated services. The underlying replication protocols of these systems execute all operations under strong consistency. However, strong consistency can hinder response time and availability, and recent replication models have turned to a hybrid of strong and relaxed consistency. This paper presents RDMA replicated data types, the first hybrid replicated data types for the RDMA network model. It presents a novel operational semantics for these types that considers three distinct categories of methods and captures their re- quired coordination, and formally proves that they preserve convergence and integrity. It implements these semantics in a system called Hamband that leverages direct remote accesses to efficiently implement the required coordination protocols. The empirical evaluation shows that Hamband outperforms the throughput of existing message-based and SMR-based implementations by more than 4x.more » « less
-
The volume of data generated and stored in contemporary global data centers is experiencing exponential growth. This rapid data growth necessitates efficient processing and anal- ysis to extract valuable business insights. In distributed data processing systems, data undergoes exchanges between the compute servers that contribute significantly to the total data processing duration in adequately large clusters, neces- sitating efficient data transport protocols. Traditionally, data transport frameworks such as JDBC and ODBC have used TCP/IP-over-Ethernet as their under- lying network protocol. Such frameworks require serializing the data into a single contiguous buffer before handing it off to the network card, primarily due to the requirement of contiguous data in TCP/IP. In OLAP use cases, this seri- alization process is costly for columnar data batches as it involves numerous memory copies that hurt data transport duration and overall data processing performance. We study the serialization overhead in the context of a widely-used columnar data format, Apache Arrow, and propose lever- aging RDMA to transport Arrow data over Infiniband in a zero-copy manner. We design and implement Thallus, an RDMA-based columnar data transport protocol for Apache Arrow based on the Thallium framework from the Mochi ecosystem, compare it with a purely Thallium RPC-based implementation, and show substantial performance improve- ments can be achieved by using RDMA for columnar data transport.more » « less
-
R-tree is a foundational data structure used in spatial databases and scientific databases. With the advancement of networks and computer architectures, in-memory data processing for R-tree in distributed systems has become a common platform. We have observed new performance challenges to process R-tree as the amount of multidimensional datasets become increasingly high. Specifically, an R-tree server can be heavily overloaded while the network and client CPU are lightly loaded, and vice versa. In this article, we present the design and implementation of Catfish, an RDMA-enabled R-tree for low latency and high throughput by adaptively utilizing the available network bandwidth and computing resources to balance the workloads between clients and servers. We design and implement two basic mechanisms of using RDMA for a client-server R-tree data processing system. First, in the fast messaging design, we use RDMA writes to send R-tree requests to the server and let server threads process R-tree requests to achieve low query latency. Second, in the RDMA offloading design, we use RDMA reads to offload tree traversal from the server to the client, which rescues the server as it is overloaded. We further develop an adaptive scheme to effectively switch an R-tree search between fast messaging and RDMA offloading, maximizing the overall performance. Our experiments show that the adaptive solution of Catfish on InfiniBand significantly outperforms R-tree that uses only fast messaging or only RDMA offloading in both latency and throughput. Catfish can also deliver up to one order of magnitude performance over the traditional schemes using TCP/IP on 1 and 40 Gbps Ethernet. We make a strong case to use RDMA to effectively balance workloads in distributed systems for low latency and high throughput.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/IA349570.2019.00009
