DaeMon: Architectural Support for Efficient Data Movement in Disaggregated Systems

TL;DR

DaeMon is a software-transparent system that significantly reduces data movement overheads in disaggregated data center architectures by employing specialized engines and adaptive data transfer techniques, improving performance and access costs.

Contribution

It introduces DaeMon, the first mechanism to transparently optimize data movement in disaggregated systems, enhancing scalability and performance across diverse workloads.

Findings

DaeMon achieves 2.39× performance improvement over page-granularity data movement.

DaeMon reduces data access costs by 3.06×.

The approach effectively handles high variability in network latency and bandwidth.

Abstract

Resource disaggregation offers a cost effective solution to resource scaling, utilization, and failure-handling in data centers by physically separating hardware devices in a server. Servers are architected as pools of processor, memory, and storage devices, organized as independent failure-isolated components interconnected by a high-bandwidth network. A critical challenge, however, is the high performance penalty of accessing data from a remote memory module over the network. Addressing this challenge is difficult as disaggregated systems have high runtime variability in network latencies/bandwidth, and page migration can significantly delay critical path cache line accesses in other pages. This paper conducts a characterization analysis on different data movement strategies in fully disaggregated systems, evaluates their performance overheads in a variety of workloads, and introduces DaeMon, the first software-transparent mechanism to significantly alleviate data movement overheads in fully disaggregated systems. First, to enable scalability to multiple hardware components in the system, we enhance each compute and memory unit with specialized engines that transparently handle data migrations. Second, to achieve high performance and provide robustness across various network, architecture and application characteristics, we implement a synergistic approach of bandwidth partitioning, link compression, decoupled data movement of multiple granularities, and adaptive granularity selection in data movements. We evaluate DaeMon in a wide variety of workloads at different network and architecture configurations using a state-of-the-art accurate simulator. DaeMon improves system performance and data access costs by 2.39$\times$ and 3.06$\times$, respectively, over the widely-adopted approach of moving data at page granularity.