Memory Disaggregation: Advances and Open Challenges

TL;DR

This paper reviews recent advances in memory disaggregation technology, discusses challenges for practical deployment, and shares insights from seven years of research on building disaggregated memory systems over ultra-fast networks.

Contribution

It provides a comprehensive overview of software challenges and open problems in memory disaggregation, based on extensive research and practical experience.

Findings

Memory disaggregation can improve resource utilization.

Ultra-fast networks are crucial for practical memory disaggregation.

Open challenges remain in hardware and software integration.

Abstract

Compute and memory are tightly coupled within each server in traditional datacenters. Large-scale datacenter operators have identified this coupling as a root cause behind fleet-wide resource underutilization and increasing Total Cost of Ownership (TCO). With the advent of ultra-fast networks and cache-coherent interfaces, memory disaggregation has emerged as a potential solution, whereby applications can leverage available memory even outside server boundaries. This paper summarizes the growing research landscape of memory disaggregation from a software perspective and introduces the challenges toward making it practical under current and future hardware trends. We also reflect on our seven-year journey in the SymbioticLab to build a comprehensive disaggregated memory system over ultra-fast networks. We conclude with some open challenges toward building next-generation memory disaggregation systems leveraging emerging cache-coherent interconnects.