Flash storage disaggregation

Abstract

PCIe-based Flash is commonly deployed to provide datacenter applications with high IO rates. However, its capacity and bandwidth are often underutilized as it is difficult to design servers with the right balance of CPU, memory and Flash resources over time and for multiple applications. This work examines Flash disaggregation as a way to deal with Flash overprovisioning. We tune remote access to Flash over commodity networks and analyze its impact on workloads sampled from real datacenter applications. We show that, while remote Flash access introduces a 20% throughput drop at the application level, disaggregation allows us to make up for these overheads through resource-efficient scale-out. Hence, we show that Flash disaggregation allows scaling CPU and Flash resources independently in a cost effective manner. We use our analysis to draw conclusions about data and control plane issues in remote storage.

Research challenges in nextgen service orchestration

“…Several recent efforts follow this disaggregated model of data centre organisation, focusing specifically on flash [87] and traditional storage [93]. While storage disaggregation is common practice (e.g.…”

“…The fog has also been propelled by the advent of programmable infrastructures, like Software-Defined Networking (SDN), Network Function Virtualization (NFV) [89,63], and data centre disaggregation [71,87,125]. These have simplified infrastructure configuration for data centre servers, storage, as well as core and edge networks.…”

HetFS: A Heterogeneous File System for Everyone

“…Nevertheless, completely replacing HDDs by more efficient SSDs can be economically prohibitive due the larger cost per capacity of the latter. More importantly, however, NAND-based SSDs have a limited number of write cycles and, in fact, recent researches on long-term SSD usage in data warehouses have proved that, after intensive usage, the SSDs degrade so much that response times may equal those of HDDs [5,6].…”

Performance Characterization of NVMe-over-Fabrics Storage Disaggregation

“…NVMe-over-Fabrics Architecture.40 times better than SATA-SSDs. Indeed, many large-scale cloud companies have reported using PCIe-based SSDs as part of their infrastructure [19,31,49,57].Generally, storage devices can either be co-located within the compute server nodes, or be placed in dedicated storage nodes accessed through the network. Large-scale cloud companies originally used scale-out infrastructure, composing their data centers out of commodity servers that tightly coupled memory, storage, and compute [3,38].…”

“…Large-scale cloud companies originally used scale-out infrastructure, composing their data centers out of commodity servers that tightly coupled memory, storage, and compute [3,38]. Unfortunately, this approach leads to inefficiencies and resource underutilization, because it fixes the ratio between compute, memory, storage, and network.Resource underutilization in data centers is a common, well-documented phenomenon [3,31,36,49]. Since the usage of different resources changes over time, predominantly independently from one another, there is no single static resource balance that fits every application that a server supports throughout its lifetime.…”