Welcome to zombieland

Nitu, Vlad; Teabe, Boris; Tchana, Alain; Isci, Canturk; Hagimont, Daniel

doi:10.1145/3190508.3190537

Cited by 41 publications

(11 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reference 8 is the first work on memory disaggregation in datacenters, extending a host memory with remote servers memory, thus decoupling CPU from memory allocations. In the same vein, several other works have studied memory disaggregation, such as Reference 1 which proposed enabling remote access to the memory of a suspended server. Not so long ago were studied fully disaggregated infrastructures, notably with storage rack disaggregation 21 and network requirements for resource disaggregation 2,22,23 .…”

Section: Related Workmentioning

confidence: 99%

“…For example, Microsoft estimates in 2020 that a 1% reduction in fragmentation within its Azure cloud platform would result in hundreds of millions of dollars in savings. This problem is fundamentally due to the classical server‐centric architecture 1 (Figure 1 top), which is massively adopted in todays DC. In this architecture, the DC is made of monolithic servers , each including all necessary hardware resources (mainly CPU, RAM, network, and disk) to run applications.…”

Section: Introductionmentioning

confidence: 99%

“…Among the approaches that have been proposed to reduce fragmentation, rack disaggregation * 1‐7 appears to be the most promising approach. It consists in consolidating a large pool of hardware resources (CPUs, RAM, disk, network, etc.)…”

Section: Introductionmentioning

confidence: 99%

“…Soft disaggregation 1,3‐6 consists in using traditional monolithic servers and adapting only the software layers to allow an application to simultaneously use resources from several machines (within a rack). The good performance offered by current DC network technologies (e.g., Infiniband) makes soft disaggregation exploitable in the current server‐centric architecture 2…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Networking in next generation disaggregated datacenters

Ekane

Tchana

Hagimont

et al. 2023

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

SummaryNowadays, datacenters lean on a computer‐centric approach based on monolithic servers which include all necessary hardware resources (mainly CPU, RAM, network, and disks) to run applications. Such an architecture comes with two main limitations: (1) difficulty to achieve full resource utilization and (2) coarse granularity for hardware maintenance. Recently, many works investigated a resource‐centric approach called disaggregated architecture where the datacenter is composed of self‐content resource boards interconnected using fast interconnection technologies, each resource board including instances of one resource type. The resource‐centric architecture allows each resource to be managed (maintenance, allocation) independently. LegoOS is the first work which studied the implications of disaggregation on the operating system, proposing to disaggregate the operating system itself. They demonstrated the suitability of this approach, considering mainly CPU and RAM resources. However, they did not study the implication of disaggregation on network resources. We reproduced a LegoOS infrastructure and extended it to support disaggregated networking. We show that networking can be disaggregated following the same principles, and that classical networking optimizations such as DMA, DDIO, or loopback can be reproduced in such an environment. Our evaluations show the viability of the approach and the potential of future disaggregated infrastructures.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Networking in next generation disaggregated datacenters

Ekane

Tchana

Hagimont

et al. 2023

Concurrency and Computation

Self Cite

View full text Add to dashboard Cite

show abstract

“…Tiered memory systems have attracted considerable research attention, with many innovative ideas being proposed across both hardware and software stacks [7,9,10,15,20,24,26,28,32,36,37,42,48]. Here, we review the most relevant research work in the context of OS support for memory disaggregation.…”

Section: Related Workmentioning

confidence: 99%

Reconsidering OS memory optimizations in the presence of disaggregated memory

Bergman

Faldu

Grot

et al. 2022

Proceedings of the 2022 ACM SIGPLAN International Symposium on Memory Management

View full text Add to dashboard Cite

Tiered memory systems introduce an additional memory level with higher-than-local-DRAM access latency and require sophisticated memory management mechanisms to achieve cost-efficiency and high performance. Recent works focus on byte-addressable tiered memory architectures which offer better performance than pure swap-based systems. We observe that adding disaggregation to a byte-addressable tiered memory architecture requires important design changes that deviate from the common techniques that target lowerlatency non-volatile memory systems. Our comprehensive analysis of real workloads shows that the high access latency to disaggregated memory undermines the utility of well-established memory management optimizations. Based on these insights, we develop HotBox -a disaggregated memory management subsystem for Linux that strives to maximize the local memory hit rate with low memory management overhead. HotBox introduces only minor changes to the Linux kernel while outperforming state-of-the-art systems on memory-intensive benchmarks by up to 2.25×. CCS Concepts:• Software and its engineering → Memory management; • Hardware → Emerging architectures.

show abstract