Related papers: Datacenter RPCs can be General and Fast
As applications become more distributed to improve user experience and offer higher availability, businesses rely on geographically dispersed datacenters that host such applications more than ever. Dedicated inter-datacenter networks have…
Remote Direct Memory Access (RDMA) is an efficient way to improve the performance of traditional client-server systems. Currently, there are two main design paradigms for RDMA-accelerated systems. The first allows the clients to directly…
Selecting the appropriate communication protocol is crucial for optimizing the performance, scalability, and user experience of web applications. In the diverse ecosystem of web technologies, various protocols like RESTful APIs, gRPC,…
Data-intensive applications in data centers, especially machine learning (ML), have made the network a bottleneck, which in turn has motivated the development of more efficient network protocols and infrastructure. For instance, remote…
High Performance Compute (HPC) clusters often produce intermediate files as part of code execution and message passing is not always possible to supply data to these cluster jobs. In these cases, I/O goes back to central distributed storage…
Today's datacenters face important challenges for providing low-latency high-quality interactive services to meet user's expectation. For improving the application throughput, recent research works have embedded application deadline…
To reproduce eScience, several challenges need to be solved: scientific workflows need to be automated; the involved software versions need to be provided in an unambiguous way; input data needs to be easily accessible; High-Performance…
We present NetReduce, a novel RDMA-compatible in-network reduction architecture to accelerate distributed DNN training. Compared to existing designs, NetReduce maintains a reliable connection between end-hosts in the Ethernet and does not…
Through the 1990s, HPC centers at national laboratories, universities, and other large sites designed distributed system architectures and software stacks that enabled extreme-scale computing. By the 2010s, these centers were eclipsed by…
High parallel framework has been proved to be very suitable for graph processing. There are various work to optimize the implementation in FPGAs, a pipeline parallel device. The key to make use of the parallel performance of FPGAs is to…
More and more latency-sensitive services and applications are being deployed into the data center. Performance can be limited by the high latency of the network interconnect. Because the conventional network stack is designed not only for…
In cloud storage systems with a large number of servers, files are typically not stored in single servers. Instead, they are split, replicated (to ensure reliability in case of server malfunction) and stored in different servers. We analyze…
Modern data centers serve workloads which are capable of exploiting parallelism. When a job parallelizes across multiple servers it will complete more quickly, but jobs receive diminishing returns from being allocated additional servers.…
We explore the feasibility of implementing a reliable, high performance, distributed storage system on a commodity computing cluster. Files are distributed across storage nodes using erasure coding with small Low-Density Parity-Check (LDPC)…
RDMA over Converged Ethernet (RoCE) has gained significant attraction for datacenter networks due to its compatibility with conventional Ethernet-based fabric. However, the RDMA protocol is efficient only on (nearly) lossless networks,…
TCP parameters in most of the operating systems are optimized for generic home and office environments having low latencies in their internal networks. However, this arrangement invariably results in a compromized network performance when…
Distributed in-memory datastores underpin cloud applications that run within a datacenter and demand high performance, strong consistency, and availability. A key feature of datastores is data replication. The data are replicated across…
Sequential computation is well understood but does not scale well with current technology. Within the next decade, systems will contain large numbers of processors with potentially thousands of processors per chip. Despite this, many…
Modern database clusters entail two levels of networks: connecting CPUs and NUMA regions inside a single server in the small and multiple servers in the large. The huge performance gap between these two types of networks used to slow down…
Elasticity is highly desirable for stream processing systems to guarantee low latency against workload dynamics, such as surges in data arrival rate and fluctuations in data distribution. Existing systems achieve elasticity following a…