Nowadays, 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 didn't 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.
Microservice and serverless computing systems open up massive versatility and opportunity to distributed and datacenter-scale computing. In the meantime, the deployments of modern datacenter resources are moving to disaggregated architectures. With the flourishing growths from both sides, we believe this is high time to write this vision paper to propose a potential research agenda to achieve efficient deployments, management, and executions of next-generation microservices on top of the emerging disaggregated datacenter architectures. In particular, we envision a critical systems research dir…
The "Disaggregated Server" concept has been proposed for datacenters where the same type server resources are aggregated in their respective pools, for example a compute pool, memory pool, network pool, and a storage pool. Each server is constructed dynamically by allocating the right amount of resources from these pools according to the workload's requirements. Modularity, higher packaging and cooling efficiencies, and higher resource utilization are among the suggested benefits. With the emergence of very large datacenters, "clouds" containing tens of thousands of servers, datacenter efficie…
Deep learning-based personalized recommendation systems are widely used for online user-facing services in production datacenters, where a large amount of hardware resources are procured and managed to reliably provide low-latency services without disruption. As the recommendation models continue to evolve and grow in size, our analysis projects that datacenters deployed with monolithic servers will spend up to 12.4x total cost of ownership (TCO) to meet the requirement of model size and complexity over the next three years. Moreover, through in-depth characterization, we reveal that the monol…
Datacenter networks routinely support the data transfers of distributed computing frameworks in the form of coflows, i.e., sets of concurrent flows related to a common task. The vast majority of the literature has focused on the problem of scheduling coflows for completion time minimization, i.e., to maximize the average rate at which coflows are dispatched in the network fabric. Modern applications, though, may generate coflows dedicated to online services and mission-critical computing tasks which have to comply with specific completion deadlines. In this paper, we introduce $\mathtt{DCoflow…
To ensure high availability, datacenter networks must rely on local fast rerouting mechanisms that allow routers to quickly react to link failures, in a fully decentralized manner. However, configuring these mechanisms to provide a high resilience against multiple failures while avoiding congestion along failover routes is algorithmically challenging, as the rerouting rules can only depend on local failure information and must be defined ahead of time. This paper presents a randomized local fast rerouting algorithm for Clos networks, the predominant datacenter topologies. Given a graph $G=(V,E…
Datacenters are the cornerstone of the big data infrastructure supporting numerous online services. The demand for interactivity, which significantly impacts user experience and provider revenue, is translated into stringent timing requirements for flows in datacenter networks. Thus low latency networking is becoming a major concern of both industry and academia. We provide a short survey of recent progress made by the networking community for low latency datacenter networks. We propose a taxonomy to categorize existing work based on four main techniques, reducing queue length, accelerating re…
The first dual-port server-centric datacenter network, FiConn, was introduced in 2009 and there are several others now in existence; however, the pool of topologies to choose from remains small. We propose a new generic construction, the stellar transformation, that dramatically increases the size of this pool by facilitating the transformation of well-studied topologies from interconnection networks, along with their networking properties and routing algorithms, into viable dual-port server-centric datacenter network topologies. We demonstrate that under our transformation, numerous interconn…
Contrary to the conclusions of a recent body of work where approximate shortest remaining processing time first (SRPT) flow scheduling is advocated for datacenter networks, this paper aims to demonstrate that per-flow fairness remains a preferable objective. We evaluate abstract queuing models by analysis and simulation to illustrate the non-optimality of SRPT under the reasonable assumptions that datacenter flows occur in batches and bursts and not, as usually assumed, individually at the instants of a Poisson process. Results for these models have significant implications for the design of b…
Responding to the "datacenter tax" and "killer microseconds" problems for datacenter applications, diverse solutions including Smart NIC-based ones have been proposed. Nonetheless, they often suffer from high overhead of communications over network and/or PCIe links. To tackle the limitations of the current solutions, this paper proposes ORCA, a holistic network and architecture co-design solution that leverages current RDMA and emerging cache-coherent off-chip interconnect technologies. Specifically, ORCA consists of four hardware and software components: (1) unified abstraction of inter- and…
Memory-compute disaggregation promises transparent elasticity, high utilization and balanced usage for resources in data centers by physically separating memory and compute into network-attached resource "blades". However, existing designs achieve performance at the cost of resource elasticity, restricting memory sharing to a single compute blade to avoid costly memory coherence traffic over the network. In this work, we show that emerging programmable network switches can enable an efficient shared memory abstraction for disaggregated architectures by placing memory management logic in the ne…
The performance of distributed and data-centric applications often critically depends on the interconnecting network. Emerging reconfigurable datacenter networks (RDCNs) are a particularly innovative approach to improve datacenter throughput. Relying on a dynamic optical topology which can be adjusted towards the workload in a demand-aware manner, RDCNs allow to exploit temporal and spatial locality in the communication pattern, and to provide topological shortcuts for frequently communicating racks. The key challenge, however, concerns how to realize demand-awareness in RDCNs in a scalable fa…
DPillar has recently been proposed as a server-centric datacenter network and is combinatorially related to (but distinct from) the well-known wrapped butterfly network. We explain the relationship between DPillar and the wrapped butterfly network before proving that the underlying graph of DPillar is a Cayley graph; hence, the datacenter network DPillar is node-symmetric. We use this symmetry property to establish a single-path routing algorithm for DPillar that computes a shortest path and has time complexity $O(k)$, where $k$ parameterizes the dimension of DPillar (we refer to the number of…
Traditional data centers are designed with a rigid architecture of fit-for-purpose servers that provision resources beyond the average workload in order to deal with occasional peaks of data. Heterogeneous data centers are pushing towards more cost-efficient architectures with better resource provisioning. In this paper we study the feasibility of using disaggregated architectures for intensive data applications, in contrast to the monolithic approach of server-oriented architectures. Particularly, we have tested a proactive network analysis system in which the workload demands are highly vari…
Distributed computing has become a common practice nowadays, where the recent focus has been given to the usage of smart networking devices with in-network computing capabilities. State-of-the-art switches with near-line rate computing and aggregation capabilities enable acceleration and improved performance for various modern applications like big data analytics and large-scale distributed and federated machine learning. In this paper, we formulate and study the theoretical algorithmic foundations of such approaches, and focus on how to deploy and use constrained in-network computing capabili…
Memory disaggregation architecture physically separates CPU and memory into independent components, which are connected via high-speed RDMA networks, greatly improving resource utilization of databases. However, such an architecture poses unique challenges to data indexing in databases due to limited RDMA semantics and near-zero computation power at memory-side. Existing indexes supporting disaggregated memory either suffer from low write performance, or require hardware modification. This paper presents Sherman, a write-optimized distributed B+Tree index on disaggregated memory that delivers …
With the increasing popularity of cloud computing, datacenters are becoming more important than ever before. A typical datacenter typically consists of a large number of homogeneous or heterogeneous servers connected by networks. Unfortunately, these servers and network equipment are often under-utilized and power hungry. To improve the utilization of hardware resources and make them power efficiency in datacenters, workload characterization and analysis is at the foundation. In this paper, we characterize and analyze the job arriving rate, arriving time, job length, power consumption, and tem…
With rapid improvements in NVM storage devices, the performance bottleneck is gradually shifting to the network, thus giving rise to the notion of "data movement wall". To reduce the amount of data movement over the network, researchers have proposed near-data computing by shipping operations and compute-extensions closer to storage devices. However, running arbitrary, user-provided extensions in a shared, disaggregated storage environment presents multiple challenges regarding safety, isolation, and performance. Instead of approaching this problem from scratch, in this work we make a case for…
Memory disaggregation has attracted great attention recently because of its benefits in efficient memory utilization and ease of management. So far, memory disaggregation research has all taken one of two approaches: building/emulating memory nodes using regular servers or building them using raw memory devices with no processing power. The former incurs higher monetary cost and faces tail latency and scalability limitations, while the latter introduces performance, security, and management problems. Server-based memory nodes and memory nodes with no processing power are two extreme approaches…
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/ba…
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/ba…
